(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
See other formats

Full text of "al-Iqtisad fil I'tiqad (Divine Predicates and their Properties) By Imam al-Ghazali"










About the Book 

In Islam, the question of the Divine Attributes has 
been treated from various perspectives. In the beginning 
it was viewed primarily as a semantic and metaphysical 
oroblem. A later introduction, its logical aspect, however, 
was its most important side. The translator attempts to 
treat the problem with this logical aspect predominantly 
in mind. Limited both by the scope of this work and by 
the task of bringing this vast material into reasonable 
compass, he confines himself to the treatment of the 
oroblem as it unfolds in Islamic Peripatetic thought, and 
the Ash'arite school, with Ghazali being the major 
representative of the latter. He tests the validity of some 
of Ghazall's statements in Iqtisdd against the former 
schools and his findings in Plato and Aristotle. 

Iqtisad is Ghazali's most sophisticated major work on 
Kalam. ' He himself holds Iqtisdd in high esteem. 


'. ffiifr- 

t "- 

_^ - |J| 

| ■£'-- 



— T" 




r— - 





^^ ■ 

*sy . 





















; > 
)■ *$ 










>\Z£%d,oU^\}\ cJ^ 


Lahore - Pakistan 






Institute of Islamic Studies, McGill University 




All rights reserved 

ISBN NO.969-432- 137-9 

First Edition, 1970 
Reprinted, 1974 
Reprinted, 1982 
Reprinted, 1990 



Muhammad 'Umar Bashir 

Jam&l Muhammad Ahmad 

Mahdl Mustafa al-H&di 

Published by Sh. Shahzad Riaz, for Sh. Muhammad Ashraf, 

Publishers & Booksellers, 7, Aibak Road, New Anarkah, 

Lahore, Pakistan 

Printed at: Ashraf Printing Press, Lahore. 



Much af what is found in Islamic philosophical 
thinking expresses conceptions similar to those held 
by other religions A historico-philosophical treat- 
ment of the problem of the divine predicates or attri- 
butes (sifdt) would not be entirely complete, therefore, 
if it ignored the conceptions of these religions, es- 
pecially given thecontention that all the sects of Islam 
were influenced in thjs respect by Judaic and Christian 
thinking. 1 Consideration of the issue of divine predi- 
cates in other religions, however, must be left to sub- 
sequent studies. 

In Islam, the question of the divine attributes has 
been treated from various perspectives. In the be- 
ginning it was viewed primarily as a semantic and 
metaphysical problem. A later introduction, its logi- 
cal aspect, however, was its most important side. 
Our attempt, therefore, will treat the problem with 
this logical aspect predomihantlv in mind. Since we 
are limited both by the scope of this work and by 
the task of bringing this vast material into reasonable 
compass, we would confine ourselves to the treatment 
of the problem as it unfolds in Islamic peripatetic 
thought, and the Ash'arite school, with Ghazaf I be- 
ing the major representative of the latter. The vali- 
dity of some of Ghazall's statements in Iqtisad would 


Al-Ghax&lt on Divine Predicates 

be tested against the former schools and our findings 
in Plato and Aristotle. Comparisons, made within the 
narrow limits we have set for this chapter, yield, we 
recognise, much discrepancy and even disharmony 
and contradiction, but we hope that the important 
doctrines will stand out clearly. 

Clearly, Muslim thinkers recognised that they had 
a problem' in the language of the Qur'an. A close 
examination of any tract on dogmatics or the here- 
siologies of al-Ash'ari, al-Baghdadi, ShahrasUral, or 
Ibn Hazm yields two aspects of this problem. The 
first aspect, the semantic one, is the description of 
God. Terms which are used to describe God are the 
same terms which are used in ordinary speech to de- 
scribe ordinary beings. In particular, the Qur'anic 
language permits the use of anthropomorphical terms 
such as eye, hand, face, etc , 2 and the present parti- 
ciple is used to designate certain aspects of the Divine 
activity. The second aspect, the ontological one, is the 
real significance of the language used to describe God. 
What is the relation between these divine attributes 
and the divine essence ? Were those attributes God's 
essence, according to His essence or superadded to His 
essence ? Definitely then, as Allard points out, the 
central problem was one of a "Revelation which God 
Himself has sent in a language that can be absorbed 

by men." 3 

Once the problem was recognised, other problems 
asserted themselves as necessary corollaries. These 
problems will be discussed presently. Some Muslim 
thinkers, however, thought they found the solution 
in the doctrine of tateU or allegorical interpretation 
of these terms used in the Qur'Sn to describe God. 


Introduction, I 

Others sought to accept these predicates and some- 
how avoid the dangers of allegorical interpretation. 
Still a third group took them literally. But this did 
not mark the end of the problem because the logical 
consequences of both denial and affirmation of these 
attributes could not be escaped. 

The group that relied most heavily on allegorical 
interpretation were the Mu'tazilah. This was neces- 
sary if their denial of the existence of real attributes 4 
in God — to safeguard His unity — were to be main- 
tained. The attributists, however, were two groups. 
There were those who accepted the real attributes 
and thus were forced to substitute the docjtrine of 
bila kayf, 5 i.e. without questioning how and without 
need for ta'wil, and those who took the terms predi- 
cated of God in the Qur'Sn literally and were thus 
abusively labelled tnus h abbihah, 6 i.e. likeners. 

We, therefore, move to examine the problem 
against three of GhazSU's assertions : (1) that God is 
existent, eternal a parte ante, eternal a parte post ; He 
is not a substance (jawhar), not a composite body and 
neither is He an accident (*arad) t nor is He defined ; 
He is not in a direction ; He is seen and is knowable ; 
God is one ■ 7 (2) that God is knowing, powerful, liv- 
ing, willing, hearing, seeing, and speaking ; 8 and (3) 
that these seven attributes are not the essence but are 
superadded to the essence. The Maker of the world is 
knowing according to knowledge, living according to 
life, powerful according to power, and so is the case 
in the rest oHhe attributes. 9 

An as yet undeclared purpose of this chapter is to 
show also that the approach and material dealt with 
in the Igtisdd is similar to the material and approach 


Al-GhazSli on Divine Predicates 

in Tahdfut al-Faldsif ah (Destruction of the Philosophers). 

In the Iqtisad we find a twofold conception of the 
unity of God. This twofold conception is also found 
in the Tahdfut and, although it is there stated as that 
of al-F&r&bl and Ibn Sina\, we know that Ghazall is 
not objecting to the concept itself as he is objecting 
to the danger involved in the methods employed to 
prove it, and hence the chiding tone in the use of 
such terms as 'old aslihim and 'aid maslakihim, i e. 
according to their premise and according to their 

The twofold conception of God's unity is stated 
by Ghaz9.ll in the Iqtisad as follows : Firstly, God's 
unity means that "God is one which means the nega- 
tion of anything other than He and the affirmation 
of His essence." t& This echoes what he says in the 
Tahdfut, viz. that the philosophers were unable "to 
prove that God is one, and that it is not permissible 
to suppose two Necessary Beings each of whom is 
without cause." 11 Secondly, the term "one," he tells 
us in the Iqtisad, means the denial of plurality in the 
sense that by "one" is meant that : 

He does not accept divisibility, i.e. He has no quantity, neithei 
definition nor magnitude. It also means that He has no equal 
in rank and absolutely no equal (Ifinid) in any manner. He has 
no contrary (did), for what is understood by contrary is that 
which alternates with a thing in the same locus, but God has no 
locus for He is without contrary. By having no equal (nid) we 
mean that all that which is other than He is created by Him, 
for if He has an equal, it would be either like Him from all as- 
pects or higher than He is or lower than He is [i.e/Ghazati ne- 
gates duality, ithnayniyyah]. God is more perfect and has. no 
equal in essence (haqiqak) or attributes. 12 

This second concept is repeated in a slightly differ* 

Introduction, I 

rent form in the Tahdfut, It is important to note again 
that, in the Tahdfut, the statements are made as those 
of the philosopher and that Ghazali's objection to 
them, we think, is one of method rather than sub- 
stance. The denial of plurality in God's essence is 
stated there as : 

Plurality comes to the essence from five aspects: (1) being 
receptive to division whether actually or conceptually ; (2) from 
the intellectual division of a thing into two different concepts, 
not quantitively, like the division of a body into matter (hayuli) 
and form (surah) ; (3) plurality through attributes by the supposi- 
tion of knowledge, power, and will, for if the existence of such 
attributes were necessary, necessary existence would be common 
to both iGod's] essence and these attributes, thereby negating 
unity ; (4) an intellectual plurality resulting from the composition 
of genus and differentia ; (5) the plurality of the essence and 
existence. 13 

The same twofold conception of God's unity is 
found in al-F&rabl and Ibn SlnS. In al-Farabl the 
terms "The First" l4 or "The First Existent" 15 (ah 
Awwal and al-Mawjud al-Awwa*) are used to designate 
"God," although the term "one" is also used. Ai- 
Farabl's conception of God's unity is stated as follows : 
Firstly, God's unity means that he is : 

(1) without any defect, that there is no existence which is 
more perfect or prior to His and there is no existence which is 
more ancient than His or on the same level, and, therefore, He 
could not possibly receive His existence from it ; He Is totally 
different by His essence from anything other than He is; (2) He is 
not dependent in His existence on another being ; (3) He has no 
contrary (did) because the existence of a being contrary to an- 
other means that they are on the same level of existence, and, 
furthermore, the perfection of the existence of any being that 
has a contrary is achieved only when that contrary is annihilated. 
God is unique '* 


Ai-Ghazilt on Divine Predicates 

Secondly, it means that : 

(I) God is not divisible in His essence as in a definition. I mean 
that He is not divisible into parts which make His substance, 
and that is because the parts of the definition of His essence 
cannot possibly point to their opposite parts in His substance. 
For if this were the case, the parts which make His substance 
would have been the cause of His existence the way the concepts 
to which the parts of the definition point constitute the cause 
of the definitum or the way matter and form cause the existence 
of that which subsists by them ; (2) if God's essence is not divis- 
ible in [the manner described above], He is even far removed 
from quantitive divisibility." 

Echoing al-Farabi, Ibn Slna's conception of God's 
unity is stated in al-Najat and al-Isharat wa al-Tan- 
bihdt to mean the following : Firstly, God's unity 
means : 

(1) it is not permissible that there be two from whom one 
necessary being is produced ; (2) there should be no multiplicity 
in the Necessary Being ; (3) the Necessary Being by His essence 
can not be a necessary being through another ... for what is made 
necessary through something else is posterior to that something 
else and depends upon it." 
Secondly, God's unity means that : 

(1) God's essence is simple in the sense that His essence has 
no principles such as the parts of the quality, parts of the de- 
finition or formula which together may cause the Necessaiy Being 
to exist whether these are like matter and form or in some other 
manner such as being parts of a formula which explains the 
meaning of His name and thereby these parts would point to a 
thin- which in existence differs essentially from another thing ; 

(2) the Necessary Being is incorporeal, not matter of bodies, not 
form of a body, not intellectual matter for an intellectual form, 
and neither is He an intellectual form in intellectual matter. 
He is not divisible into quantity or principle or in definition 
(qmwl)* 9 

This concept of the unity of God is an elaboration 


Introduction, I 

oh the embryonic Qur'Snic principle "Naught is there 
anything like Him" (laysa kamithiihi shay'), and both 
Muslim philosophers and theologians strived to re- 
concile this principle with other obviously anthropo- 
morphic terms. The methods, however, differed. Now, 
by combining this Qur'anic principle with the philoso- 
phical principle of the incorporeality of God, the 
three men arrived at the idea that God is indefinable 
and indivisible which means the exclusion of the genus 
and differentia. How did they all arrive at this con- 
cept of absolute simplicity of God. a concept which 
seems to be responsible for the rejection of the attri- 
butes by al-Farabi, Ibn Sina, and the Mu'tazilah? 

We should now turn to Plato and Aristotle as the 
latter is interpreted by al-Farabl and see how the prob- 
lem is dealt with in their works, for both al-Far&bi 
and Ibn Sina base themselves on them in this respect 
and in the case of al-Farabi, at least, there is an at- 
tempt to try to bridge the gap between Plato and 
Aristotle which has been epitomised in aUJam* bayn 
ra'yey al-Hakimayn. 

In his Dialogues Plato treats God as either one of 
the ideas or as a supreme idea and, therefore, only 
through the understanding of that treatment can we 
reconstruct Plato's notion of God. 

In the Phaedo, Plato describes the ideas as 
"simple" and "unchanging." 20 In the Republic they 
are said to exist each as a unity. 21 In the Parmenides, 
Socrates, Plato's spokesman, tells of a distinction 
between the ideas and the things that partake ^of 
them. 22 In the Sophist, the ideas are described as in- 
corporeal. 23 In the Timaeus, the idea is said to be : 
Immovably the same forever; it cannot become younger or 


Al-Ghatill oh Divine Predicates 

older by time, nor can it be said t hat it came into being in the 
post, or has come into being now, or will come into being in the 
future; nor is it subject at all to any of those states which 
affect moving and sensible things and of which generation is the 
cause. 24 

Along with such predicates as simple, unchanging, 
existent, forever the same, and incorporeal, which 
Plato uses to describe God, he also uses such predi- 
cates as good 25 and "wise ... for that is a great name 
which belongs to God alone." 2 * How does Plato re- 
concile such predication with his conception of the 
simplicity of the ideas ? 

In the Parmenides, Plato says : 

But tell me, Zeno, do you not further think that there is 
an idea of likeness, detached and existing by itself, and an op- 
posite idea, which is the essence of unlikeness, and that in these 
two you and I and all other things to which we apply the term 
may participate — things which participate in likeness become 
in that degree and manner like; and so far as they participate 
in unlikeness become in that degree unlike; or again are both 
like and unlike-in the degree in which they participate in both? 
And even though all things partake of both opposites, and be 
both like and unlike to themselves by reason of this partici- 
pation, where Is the wonder? 2 ? 

It is clear then that Plato does not object to describ- 
ing sensible objects by opposite terms such as likeness 
and unlikeness, because sensible objects are composite. 
But of the ideas which, we are told, are simple, it 
seems, no opposite term should be predicated be- 
cause ideas do not at any time change. This is so, for 
clearly, as Plato says : 

The same thing cannot act or be acted upon in the same 
part or in relation to the same thing at the same time, in con- 
trary ways ; and, therefore, whenever this contradiction occurs 
in things apparently the same, we know that they are really not 


Introduction, I 

the same, but different. 29 

But since Plato himself has predicated certain terms 
of the ideas, what are these terms to be considered ? 
Plato talks of ideas which are capable of "intermix- 
ture" 29 and others which are not, We can, therefore, 
conclude that Plato does think that there are terms 
which can be predicated of God who is either one of 
the ideas or a being above the ideas. 

In a chapter entitled "al-Muthul al~Afla\un\yyah 
wa mawqif Aristii tninha,"* al-Farabi points out that 
Aristotle stood firmly opposed to Plato's incorporeal 
beings which are the ideas or, as al-Farabi describes 
them, "the divine ideas". Excepting this, in his 
attempt at harmonising the thoughts of Plato and 
Aristotle, al-Farabi mentions that in his Athdlujiya 
(Theology), Aristotle affirms the existence of spiri- 
tual forms" which are in God. God, according to al- 
Farabi's interpretation of Aristotle, is described as 
unlike anything in His quiddity (anniyyatihi), as indi- 
visible, immutable, and unalterable. 51 Aristotle per- 
mits the description of God by terms which are 
natural, logical, and univocal. 32 But when these terms 
are applied to God or to the "spiritual forms/' i.e. 
the ideas, they are used in a manner different from 
their application to ordinary beings. What al-F&r&bl 
here calls natural terms (alfaz tab? iy yah) are those 
terms which Plato calls "not opposite" as in describ- 
ing man, for example, by "animal" and "rational". 

Although we know now, on al-Farabi's authority, 
that Aristotle permits predicating univocal terms 
(alfaz mutawati'ah) of God, we should seek confirma- 
tion of this assertion in Aristotle. Does Aristotle's in- 
divisibility of God exclude the genus and differentia 


Al-Ghazali on Divine Predicates 

as is clearly the case with Gbazall, Ibn Sina, and al- 
Firabi ? Aristotle enumerates four predicables which 
are either genera or a species or properties or acci- 
dents. 33 What of these four is to be predicated of God 
and who is said to be one and incorporeal by Aris- 
totle ? Through the understanding of Aristotle's struc- 
ture of <the one, we expect to arrive at an answer. 
Now, accidents must be eliminated for they are to be 
found only in corporeal beings. According to Aristotle, 
the term "one" has many meanings: 

for in general those things that do not admit of division are 
called one in so far as they do not admit of it, e.g. if two things 
are indistinguishable qua man, they are one kind of man; if 
qua animal, one kind of animal; if qua magnitude, one kind of 
magnitude. 34 

The question is whether Aristotle considers God as 
these beings which could be called one even though 
they are divisible into quantity, substance, and acci- 
dent. Some things are called one because of the unity 
of their definition, e.g. man, although every definition 
is divisible into genus and differentia. Aristotle predi- 
cates many terms of God such as the "one," 33 eternal 
and unchangeable, 30 the good and the beautiful, 37 
causing motion, 38 and thinking. 39 As al-Farabl points 
out, such predicates are used by Aristotle to describe 
God in a manner different from their application to 
corporeal beings. Things which are called one are 
divisible but God as an incorporeal being is indivis- 
ible. Things which are described as moving are moved 
bat God is a prime mover ; things which are called 
thinking think of an object external to themselves 
but God, when He thinks, "thought arid object of 
thought are the same." 40 These predicates must fall 



Introdudton, I 

within Aristotle's previously mentioned four predi- 
cables. God's indivisibility excludes parts or divisi- 
bility into substance and accident or matter and form. 
In the Topica, however, Aristotle argues that genera 
and species have no real existence and hence they 
could not be prior to the subject defined, 41 or be its 
cause. 42 What causes one species to break up into 
numerically many individuals is matter but the "pri- 
mary essence has no matter, for it is complete real- 
ity. So the unmovable first mover (God) is one both 
in definition and in number." 43 Now we realise that 
Aristotle includes under "immovable mover" the 
heavenly spheres, but these are themselves moved by 
an accidental motion by something else," 44 while God 
is absolutely immovable. And in any case these prin- 
ciples are immaterial and hence the differences be- 
tween them are specific differences, i.e. each of them 
is a unique species although their genus is com non. 
But we have pointed out that Aristotle reminds us 
that genera and species have no real existence 45 and, 
therefore, their predication of God does not introduce 
real division in His essence. The argument that God's 
case, then, is not different from that of any difinitum, 
for a definition implies a distinction between matter 
and form, despite Aristotle's assertion that God is 
immaterial, would be answered by him that matter is 
not necessarily sensible. It is also intelligible 46 

Ibn Slna, on the other hand, agrees with Aristotle 
when he says that the principle mover cannot be 
many. The other heavenly spheres, also, are movable. 
The movable mover, however, is numerically "one" 
and he moves the rest of the kurat al-samd,' 41 i e. the 
heavenly spheres which are immaterial 48 (mufariqah 


AlGhazdU on Divine Predicates 

lilrtnaddah). He also affirms Aristotle's idea that the 
genus and species which are universal have no real 
existence {la kuUl 'ammiyfil wujud) : 
no universal is general in existence, for the existence of a 
universal in actuality is only in the intellect and that is the 
form which exists in the intellect, whose actual or potential 
relation'to every single thing is the same* 
In the logic of the NajSt. Ibn Slna classifies five uni- 
versal predicables, four of which we have already 
shown cannot be predicated of God in the ordinary 
sense These five predicates (at-alfdz al- kkamsah) are : 
genera, species, differentiae, property (khassah), and 
the accidents. 30 Of these, property is rather problem- 
atic Starting with the statement that the First 
Necessary Being is simple, indivisible, and free from 
matter and its attachments. 51 Ibn Slna argues that 
the divine predicates are properties which, when 
applied to God, imply no multiplicity or divisibility 
of the essence of the "one".* But Ibn Sioa predicates 
them in a special manner : God, for example is will- 
ing, knowing, living, powerful, li *****>" *? m 
virtue of His essence. This is so, says Ibn Slna, for, 
when it is said that God is Intellect, is intellected, and in- 
tellects, this means in actuality God's existence, from which 
matter and its attachments are excluded and a certain relation ts 
posited. Therefore, when we say the First, we only mean the 
relation of His existence to the whole.** 
This echoes Aristotle's statement : 
And life also belongs to God; for the actuality of thought is 
hfe and God is thaf actuality ; and God's -f^^^t 
ity is life most good and eternal. We say, therefore, that God is 
a living being, eternal, most good, so that life and duration 
continuous and eternal belong to God ; for this « God » 

Explaining this further, Ibn Slna asserts that 


Introduction, I 

ordinary beings have life because of two faculties 
which make life possible for them ; they will, because 
of an intention, but His knowledge is His will 56 for 
the reality which God intellects is His very know- 
ledge, power and will 37 because what issues from God 
issues in virtue of His essence (li dhdtihi) and does not 
entail divisibility or multiplicity in His essence for 
the *' First Cause" is numerically one, is immaterial 
and pure intellect. 38 What Ibn Slna meant by "a 
certain relation" 39 could be interpreted as what is 
known as positive predicates or attributes of action, 
for he also refers to what he calls "Sifatrna'aas-salb" 60 . 
or negative attributes. He explains this kind of pre- 
dicates as follows : "if someone describes The First as 
'substance,' he would mean His existence with the 
negation of His being in a substratum and if he 
describes Him as The One, he would mean the Exist- 
ent Himself with the negation of His divisibility into 
quantity, the negation of definition, and the negation 
of a partner (sharik) to him." 61 This is Ibn Sina's 
position as we can see and as stated in al~Najat. 

The Ash'arites are represented by Ghazall. They 
reject the views of Ibn Sin5, viz. that the attributes 
are the essence, for, according to Ghazall, this would 
be tautological, 6 i.e. tantamount to saying God is 
God. We have already quoted Ghazali as saying that 
God's attributes are eternal and superadded Jo the 
essence, 63 and he forewarns us that these atiributes 
are not the essence but rather point to the essence. 
In the fahafiU, however, he attacks both al-Farabi 
and Ibn Sina and their arbitrary conception of the 
simplicity of God c4 which forced them to deny to God 
real attributes. This arbitrariness, Ghazali points out 


Al-Gkaxail on Dtvtne PtedtcaUs 

stems from their doctrine of Wtjib al-Wujud, ^Neces- 
sary Being" which they have acquired from their 
conception of the necessary and the possible*** not 
from the distinction between the eternal and the con- 
tingent. Ghazali's understanding of the doctrine of the 
Necessary Being is that it furnishes the proof for the 
existence of a being who is without a cause but not 
without composition if their conception of the unity 
of the essence rejects the multiplicity of attributes £ 
which are eternal. He urges the philosophers to set 
aside the doctrine of the Necessary Being and the 
doctrine of simplicity because it is vulnerable" and 
to see whether the attributes are incompatible with a 
being who has no cause He argues that there is no 
incompatibility, a conclusion he draws from his 
enumeration of three types of relati on* • t hat may 
exist between God's essence and the attributes («) 
both essence and attributes are independent rom 
each other ; (b) they depend on each other ; (c) the 
essence is not dependent on the attributes but the 
latter are. His choice is the third kind of relationship. 
Thus he makes by providing his own conception of 
the expression Necessary Being which he explains as 
meaning having no efficient cause," and -just as .the 
mind is capable of intellecting an eternal Existent 
who has no cause, it is also capable of "Electing an 
eternal Existent who has attributes and who has no 
cause for the existence of both His essence and His 

at 1Sfclcludes his discussion of the problem in 
his usual uncompromising tone and in scathing terms 
by belittling the intellectual methods and proofs of 


Introduction, I 

1. It is the consensus of Western scholarship that this is the case. See 
particularly Wolf son, "The Muslim Attributes and the Christian Trinity," 
Harvard Theological Revi$w, 49 (1956), 1-18. 

2. Iff. Allard, Le problem* des attributs divins, Imprimerie catholiqwe. 
Beyrouth. 1965, p. 14. 

3. Ibid. 

4. Sbahrastani, al-Milat wa al-Nikal, Leipzig, Otto Harrasaowite, 1923. 

p. 30. 

5. Ibid, pp. 65-6. 

6. Ibid , p. 65, 75-6. They are also known as the eorporealistt {Muiassi- 
tnah). See Maqatdt of al-Ash'ari. Hitter. 1963, p. 207. 

7. Ghazaii. Iqtisdd Fil-J'tiqdd, Ankara, 1V62, p. 4 

8. Ibid., p. 79. 

9. Ibid., p. 129. 

10. Ibid., p. 73. 

11. tahifut al-Fatisifak, problem V, ed. Sulaymftn Dunya, Cairo, 1947, 
p. 133 ; cf. Maqtsid al-Faldsifah , where Ghazaii enumerates twelve points 
concerning the Necessary Being, ed. Dunya, Cairo, 1961. 

12. Iqtisdd. pp. 73-5. 

13. Tahd/ut al-Faldsifah, V, pp. 136-9. 

14. Abft Nasr al-Farabi, Kitdb as-Siydsah al-Madaniyyah, ed. F M. 
Najjar, Beirut, 1964. p. 42j 

15. Kitdb A'rd' ahl al Madinah aUF&dilah, 2nd ed., ed. al-Kurdi. Cairo, 
1948, p. 2. 

16. (a) As-Siydsah al-Madaniyyah, pp. 42-3. 
{b) A'rd' ahl at- Madinah, pp. 4-5. 

17. (a) As-Siydsah al-Madaniyyah, p. 44. 
{b) A'rd' ahl at- Madinah, p. 8. 

18. Ibn Sina' dl-Najdt Fi al-Hihmah al-Manfiqiyyah wa al-Tabi'ah at- 
Udhiyyah, ed. al-Kurdi. Cairo. 1938, Part II, p. 227. 

19. (a) al-Nafdt, pp. 227-8. 

(6) al-Ishirdt, p. 278, where Ibn Sin* says '. that the Necessary 
Balng participates with nothing in any generic or specific concept, that His 
essence has no definition because i\ has no genus nor differentia. See Tusi's 
commentary, pp. 472-3. 

20. Pkaedo. 432-78d. 

21. Rep., VI. 369-507. 

22. Parm., 673-1 30b. 

23. Soph., 398-24$b. 

24. Tim., 723-4-38a. 

25. Ibid.. 717-29e. 

26. Phaedrus, 188-278d. 

27. Parm , 671 -2-1 29a 


Al-Ghatitt on Divint Preiitafs 
28. Rtp.. IV, 28»-M6b. 

.^ w2!tM6s — An.to«..«y... VIII. .0. 2«7b. 84. D. O*. 
19, 279a, 19-21. 

32. Ibid, p. 107. 

33. Toxica, 1, 101b, 17-25. 

34. Metaphysica, 1016b. 3-7. 

35. Ibid., 987a. 20-3. 

36. Ibid , 987b, 16-7. 

37. Ibid., 1078a, 31. 

38. Ibid , 1072b. 1-3. 

39. Ibid.. 1072b. 18-22. 

40. Ibid.. 1072b, 22. 

41. Toptca, Vl-r. 141a, 26-7, 

42. Anal, post., 11-10, 93b. 38-29. 

43. Metaphysica, XII-1. 1074a, 35-7. 
44 Phys . VIII-6. 259b, 29. 

45. Metaphysica, XII-5, 1071a., 19. 

46. Ibid., VIII-6. 1045a, 34. 

47. aI'Najdt,&266. 

48. Ibid., p. 267. 

49. Ibid., p. 221. 

50. Ibid., p. 8 et passim. 

51. Ibid., p. 244. 

52. Ibid., pp. 249-251. 

53. Ibid., p. 250. 

54. Ibid., p. 251. 

55. Aristotle, Metaphysica, XII-7, 1072b, 26-8. 

56. al-NSjat, p. 250. 

57. Ibid., p. 274. 

58. Ibid , pp. 274-5 

59. See note 54. 

60. al-Najat, p. 251. 

61. Ibid. 

62. Iqlisad, p. 132. 

63. See notes 7, 8. 9. 

64. Tahafut al-Falasifah, VI, p. 147. 

65. Ibid., pp. 147-50. 

66. Ibid., p. 147. 

67. Ibid., p. 149. 

68. Ibid., p. 148. 

69. Ibid., p. 150. 

70. Ibid. 




Although Gbazali may have hurled criticism 
against kaldtn, he retained a mutakallim side to his 
complex person. As a mutakallim, he dealt with the 
major issues of kaldm which al-Ash'ari had treated 
and defended, except that at Ghaz&ll's hands these 
issues receive a fresh treatment, more thorough, more 
comprehensive, and more systematic than in any of 
his predecessors. Particularly new is his treatment of 
these issues as problems of logical judgment. Ghazall 
was the first mutakallim to use Aristotelian logic to 
defend matters of dogma, a method which al-Juwayni 
had attempted without exhausting its potentials. 1 
This "new" method is abundantly demonstrated in 
Ghazall's Tahafut al-Falasifah and al-Radd 'aid al- 
Bdtiniyyah, both of which he considers tracts on dog- 
matics. 2 It appears also in QawaHd al-Adiilah and aU 
Risalah al-Qudsiyyah and in this work with which we 
are dealing, al-Iqtisdd. 

In defending the use of logic to prove issues of a 
dogmatic nature, Ghazall says : 

But logic is not a monopoly of theirs [i.e. the philosophers]; 
rather it is the principle (asl) which we call in the art ol kaldm 
the "Book of Lo Jcal Reasoning" (Kitab al-Naxar). They altered 
its expressions to sour.d impressive. We may also call it the 
"Book of Dialectics" {Jadal) or we may call it intellectual per- 
ception. When a dilettante hears the term "logic." he thinks it 
is a mysterious art which the mutakallimun do not know and 


Al-GkazSlt on Divine Predicates 

only philosophers have access to it. 3 

Numerous examples can be given to show Ghaz&ll's 
complete acceptance of logic as a sound method for 
acquiring truth. He says in Afaqasid al-Faldsifah : 
... as for logical judgments, they are mostly correct and are 
rarely erroneous, but they fi.e. the philosophers] differ from the 
People of the Truth concerning them only in terminology and ap- 
plications but not in meaning and conclusions. 4 

This final acceptance of Aristotelian logic is exempli- 
fied in Iqtisat by Ghazall 's use of categorical syllo- 
gisms as distinct from hypothetical syllogisms which 
are certain truth-functional arguments. It may be said 
that what are traditionally known as syllogisms are 
arguments wherein a categorical statement is derived 
as a conclusion from two categorical statements as pre- 
mises, the three statements being so related that there 
are altogether only three terms, each of which appears 
in two of the statements. Ghaz3.II uses what he calls 
valid syllogisms. A valid syllogism is said to be of such 
form as to be incapable of leading from true premises 
to a false conclusion. Two examples of his syllogisms 
run as follows : 

(a) Every originated being requires a cause for 
its origination, 

the world is originated, 
therefore the world bas a cause. 5 

(b) Any masterly work proceeds from a powerful 

the world is a masterly work, 

therefore the world proceeds from a powerful 

Then Ghazall explains the terminology used and at 
the end offers the demonstration. Each premise is re- 


Introduction, II 

peated and proved. In the major premise of example 
(a), after pointing out that existence is fundamentally 
real, Ghazali admits the existence of beings such as 
bodies and accidents, which can be perceived "by the 
senses," 7 and its reality does not have to be demon- 
strated. But the major premise also includes the exist- 
ence of a being (i.e. a cause which is not perceived 
by the senses). It exists, none the less, and the world 
exists through it and through its power. 8 To prove 
this latter assertion Ghaz&U merely explains the mean- 
ing of the terms "cause" and "being which requires 
a cause," and "if the opponent understands the terms, 
his intellect would necessarily accept [the explana- 
tion]." 9 He then deals with the minor premise, "the 
world is originated". This is not perceived by the 
senses and hence must be proven. 10 After pointing out 
that by the world he means nothing other than com- 
posite bodies, he develops his demonstration in the 
form of another syllogism : 

No body is free from originated things ; 

all bodies which are not free from originated 


are originated ; 

therefore every body is originated." 
Ghazall dwells at length on the major premise which 
he proves in two sections. The first section is that move- 
ment and rest are real things and "the substance is 
never free from them." ,a The second section is that 
both movement and rest are originated things. As for 
movement and rest, Ghazall says that this is a super- 
fluous issue for a book like Iqtisdd and that it does 
not deserve all the attention it received in books on 
Kaldm, 13 for no sane person would d*>ubt the existence 


Al-GhazSll on Divine Predicates 

of accidents in himself, such as pain, sickness, hunger, 
thirst, and other states. The observation of the exter- 
nal world would also provide similar findings, viz. 
that everything is in constant change and that such 
change is contingent. But since his opponents, con- 
cerning this point, are the philosophers, Ghaz3.li feels 
he should discuss it because they do not concede the 
contingency of the world although they concede that 
the bodies in the world are divided into (a) heavenly 
spheres which are perpetually in motion, their indi- 
vidual motions being originated though they are eter- 
nally and sempiternally in succession, and (b) the 
four elements which exist in the sublunar world and 
which share an eternal matter that gives subsistence 
to their forms and accidents; and the latter two being 
originated. The four elements, however, are never 
free from originated forms because from their origi- 
nated mixtures are formed the elements, plants, and 
animated beings u Ghazali says that a lengthy dis- 
cussion is not necessary to prove what is self-evident 
but for form's (rasm's) sake he would discuss these 
issues. Proving what is self-evident obscures issues 
rather than sheds light on them. 15 He realises that 
what he calls self-evident meets with two objections. 
He spends some time explaining the second of these 
objections through the explanation of the expression, 
"an accident which subsists in a locus". This way, 
Ghazali feels he has clearly demonstrated the major 
premise of the second syllogism, "the world is origi- 
nated," for it is precisely against the philosopher's 
assertion that the world is not originated, that 
Ghazali states the minor premise of his second syllo- 
gism and it is on this that the whole thesis rests, and 


Introduction, II 

it is here that a real demonstration appears and it is 
an argumentutn ad absurdum. As a matter of fact, the 
conclusion of Ghazali is that "the truthfulness of our 
premise has thus become evident by means of the 
third method of reasoning mentioned in the fourth 
preliminary at the beginning of the book." 16 The 
third method in the fourth preliminary states that the 
opponent is free to contest the validity of the conclu- 
sion by refusing to admit the premise." This freedom 
which he grants to the opponent is "honest, but 
reflects fragility in his rational argument." 18 For 
Ghazali, no discrepancy exists in so far as the oppo- 
nent would admit the reality of the origination of the 
world, a postulate necessary for the very groundwork 
of Ghazalfs rational system. This postulate, he 
argues, is known through reason without any recourse 
to the revealed word {shar 1 ) being necessary. 19 But, 
implicitly, it is shar' which imposes on reason the 
truthfulness of the postulate. Ghazali concludes his 
demonstration of the origination of the world by 
various verses from the Qur'an because for him that 
is all self-evident 20 and if he demomtrates' what is 
self-evident, it is for the sake of logic and out of sheer 
desire to consolidate the system he is defending. In 
ending his first treatise by the Qur'anic verse: "if 
there were gods other than Allah in them [i.e. in 
heavens and earth], they would be decaying: 21 This, 
for Ghazali, is the unanswerable proof and one should 
not seek further demonstration. This is not mere 
literary analysis or mere impression. 22 It reflects the 
conviction of Ghazali's inner self. Although he argues 
that even prophets were not sent to prove the exist- 
ence of God and the origination of the world, but 


Al-Ghazali on Divine Predicates 

were sent to teach His unity, we find elsewhere that 
he permits arguing for the existence of God when the 
arguments are drawn from and sustained by the 
Qur'an." Such demonstration, he concedes, is direct- 
ed toward the mass of believers. In the Qistds, 24 the 
same verse is presented as a model of a conditional 
conjunctive syllogism or mtzdn at-taldzurn. All this of 
course is a way of showing that reason alone cannot 
reach absolute certainty, for reason can help only in 
taking positions vis-a-vis the revealed word of God. 
For Ghaz&K, then, reason must be used to reach a 
golden mean or middle course in belief, not absolute 
certainty. This is clear in his arguments in the seventh, 
eighth, and ninth thesis of the first part, 25 and the 
second and third parts on attributes and their proper- 
ties. 26 There Gbazali's sentences become longer, inflat- 
ed, and his satisfaction becomes apparent. But even 
here, the explanation of the terms takes precedence 
over the demonstration, and Ghazall asserts with con- 
fidence what he believes to be the golden mean in 
belief or the middle course which happens to be that 
of the Ash'arites. 27 

The importance of tqtisdd lies in this extensive use 
of syllogistic logic and "the attention to objections 
from a neo-Platonic standpoint," 2I both of which can 
be seen in the proofs of the existence of God and the 
visio beatifica, which in Montgomery Watt's judgment 
are in fact Ghaz&U's greatest contribution to the 
latter development of Islamic theology and accentu- 
ate the fact that what Ghazall did was to "effect a 
complete fusion of the Greek and Islamic intellectual 
traditions." 29 

Iqtisdd is Ghazall's most sophisticated major work 


introduction, 11 

on kaldm. Its full name, al-lqtisdd Vxl-Viiqdd, has 
been variously translated, but the most appropriate 
translation is The Golden Mean in Belief. Ghazall 
himself holds Iqtisdd in high esteem and, whenever 
the occasion calls for the proofs of the creed, he refers 
to it. His own evaluation of the book can be found in 
many of his works. Twice in the Ihya', Ghazall men- 
tions Iqttsdd as the book containing more adequate 
proofs than are normally found in other works of the 
theologians, and as a work which reveals the essence 
of the science of kaldm. 30 In Ghazali's own words : 
As for the proofs of the creed we have set them in al-Risdlah 
al-Qudsiyyah in about twenty pages and it is a chapter of the 
Kitib Qawd'id al-Adtllah of the lhy&' and as for their proofs, with 
more thoroughness and sophistication in raising questions and 
objections, we have set them in the Golden Mean in Belief, in 
some hundred pages. It is a book devoted entirely to the very 
essence of the science of the theologians, but it is more adequate 
fn its proofs and more apt to knock at the doors of knowledge 
than the scholastic jargon which is normally met in the works 
of the theologians. 31 

Although Ghazili does not mind that the mass of 
the believers may read it, he still considers it his 
single most important statement on the creed. Dis- 
cussing Kaldm in the Jawahir, he says that he has 
written two treatises on kaldm explaining the creed 
"on two levels, calling the lower level al-Risdlah 
al-Qudsiyyah, and the higher level al-lqtisdd Fil- 
Vtiqdd."* 2 


1. Hanna al-Fakuri and Khalil al-Jarra, TAtthh al-Fmlsaha at- 
Arabiyyah. Dfif al-Ma'Srif, Beirut, 1957. Vol. II, p. 267. 

2. Gbazili, Jow&hir al-Qur'Sn, Kurdistan Pres», Cairo, 1329 a.m., 1st ed. 

p. 26. 


Al-GhaziU on Divine Predicates 

3. Tahdfutal-Fal&sifah, ed. S. Duniya, Cairo, 1947, p. 45. 

4. Maqifm al-Faldsifah, Logic, ed S Doniya, Dar al-Ma'arif , Cairo, 
1961, p. 32. 

5. Iqtisad, p. 24. 

6. Ibid , p. 30 Our translation Chapter I, p. 1. 

7. Ibid. 8. Ibid., p. 25. 
9. Ibid. 10. Ibid., p. 26. 

11. Ibid. 12. Ibid. 

13. Ibid., p. 27. 14. Ibid. 

15. Ibid., p. 28. 16. Ibid., p. 34. 

17. Ibid., p. 17. 

18. Farid Jabre, La notion de certitude selon Ghazili, Librairie philoso- 
phique, Paris, 1958, p. 83. 

19. Iqtisad, pp 210-3. 20. Ibid., p. 79. 

21. Ibid. 22. Farid Jabre, op. cit. p. 83. 

23. Iljam al-'Awdm, p. 27 ; F. Jabre, op. cit., p. 83. 

24. Al-Qis(&s al-Mustaqim, ed. Mustafa al-Qabbani ad-Dimaahqi, 
Matba'at at-Taraqqi, Cairo, 1900, 1st ed., pp. 50-1. 

25. Igtisdd, pp. 42 ft. 

26. Ibid., pp. 79 ff., see our translation Chapter I, p. 1, and Chapters 
VII, VIII. 27. Ibid., pp. 129-33. 

28. W. Montgomery Watt, Muslim Intellectual, A Study of al-Ghazali, 
Edinburgh University Press, 1963, p. 123. 

29. Ibid. 

30. IHyd' 'Ulum ad- Din, Lajnat Nashr al-Thaqafah al-Islfimiyyab, Cairo, 
1356 A.H., Vol. I, pp. 88. 169. 

31. Al-Arba in ft Usui ad-Din, p. 25, cited by Dunya, p. 128, and Watt, 
p. 119. 32. Jawahir al-Qur'an, p. 25. 

A Note on the Footnote System 

The reader will encounter in the text of the translation two 
systems of footnotes. When the number is followed by a letter 
"a" as in 5a, for example, such a number refers to a variant 
reading to be found in the Appendix. When a number is not 
followed by a letter, the reference is to a quoted Work, the ex- 
planation of a term or an idea of a general elucidation of the 
text, in other words a regular footnote. 

In the Appendix, however, there are three sets of two letters 
each, AK, AS and BA which we are using to refer to the three 
manuscripts utilised In preparing this translation. When, for 
example, the symbol AK-30 appears in the Appendix, the num- 
ber 30 refers to the folio number of the particular manusciipt. 



It is my great pleasure to express, in some measure, 
my deep indebtedness to the able and effective guid- 
ance of Professor Toshihiko Izutsu, formerly of the 
Institute of Cultural and Linguistic Studies at Keio 
University, Japan, and presently Professor of Philoso- 
phy at McGnTs Institute of Islamic Studies, Tehran 
Branch, and the constant encouragement and concern 
of Professor Charles J. Adams, Director of the Institute 
of Islamic Studies at McGill University, who, indeed, 
created the very conditions for my work and have in 
part read the original manuscript of this translation . 
It is my pleasant duty to record my thanks to Mr 
Paul McClean who gave so much of his time to look 
into the early chapters of the translation and suggest 
valuable alterations, and to Miss Karen Lee Koning 
and Mr Rusen Sezer who went through the tiresome 
task of reading the final proofs. 

Although in preparing this text I have used the 
manuscripts referred to in the Appendix, the 1962 
Ankara edition of Iqtisad of Professors Hussayn Attay 
and Ibrahim Cubukou of the Ankara Universitesi, 
Ilahiyat Facultesi, has been very helpful, and I am 
grateful for their allowing me its use. 



Dedication, v 

I. The Problem in Its Historico-Philosophical Context, vii 
II. Al-Iqtisad fit-I'tiqad, Method and Importance, xxiii 
A Note on Footnote System, xxx 
Acknowledgments, xxxi 

I. The Second Pivotal Point concerning the Attributes, 1 
II. The Second Attribute : Knowledge {'Urn). 25 
III. The Third Attribute : Life, 29 
IV. The Fourth Attribute : Will, 30 
V. The Fifth and Sixth Attributes concerning Seeing and 
Hearing, 40 
VI. The Seventh Attribute concerning Speech, 47 
VII. The Second Section of this Cardinal Point, 65 
VIII. The Second Property of the Attributes. 77 
IX. The Third Property, 80 
X. The Fourth Property, 98 

Appendix, 102 
Btbliographv, 109 
Index, 112 


Chapter I 


It contains seven assertions. We assert that God 
is knowing, powerful, living, willing, hearing, seeing, 
and speaking. These are seven attributes from whieh 
the consideration of two matters arises. One of them 
is what is proper to each attribute separately, and the 
second is what is common to all the attributes. Let 
us begin with the first point, viz. the establishment of 
the basis of the particular 1 ** attributes and the 
explanation of their special particular properties. 

The first attribute is Power {qudrah). 1 We assert 
that the Creator 2 ' of the world is Powerful because 
the world is a masterly work, well ordered, perfectly 
arranged, including varieties of wonders and signs, 
and all that points to power. 

Properly to order the syllogism, 2 we say : 

Any masterly work proceeds from a powerful agent. 
The world is a masterly work. Therefore, it proceeds 
from a powerful agent. Concerning which of the two 
premises is the dispute ? 

If it be asked: 

Whydid you say that the world is a masterly work ? 

We answer: 

We mean by being "masterly," its perfect order, 
systematic arrangement, and symmetry. He who ex- 
amines closely the members of his own body, external 

•For numbers with a letter, please see the Appendix at the end of the 
translation. Footnotes are given at the end of each chapter. 


Al-Ghazali on Divine Predicates 

and internal, ** will perceive wonders of perfection 
which surpass accounting. This is a premise the source 
of the knowledge of which are the senses 4 * and ob- 
servation ; and, thus, it cannot be denied. 

// it is said : 

Then, how did you know the other premise, viz. 
that every masterly and well-ordered work proceeds 
from a powerful agent ? 

We say : 

This is attained through rational 5 * necessity. The 
intellect confirms this without an apodictic proof, and 
the intelligent person will be unable to deny it. Even 
though this is the case, we shall furnish a proof that 
will cut the ground from under denial and obstinacy. 

We say : 

We mean by His being powerful that the act pro- 
ceeding from Him either proceeds from Him through 
His essence or through a concept *• superadded to 
His essence/and it is absurd that it proceeds from Him 
through His essence, because if this were the case the 
act would be eternal a parte ante (qadim) along with 
the essence (dhat)? This proves that it proceeds from 
something superadded to His essence [dhdt). We call 
this superadded attribute, through which the act 
exists, 7 ' power, since the proper meaning of the term 
"power" (qudrah) in [the Arabic] language is nothing 
other than the attribute by which the act is made pos- 
sible for the agent, and through which the act occurs. 
This description is proven by the decisive distinction 
which we have mentioned, and by power we mean noth- 
ing other than this attribute which wehave established, 

// it is said : 

But this [argument], regarding power, could be 


The Second Pivotal Point concerning the Attributes 

turned against you, for, since power is eternal 
(qadimah), why 8 " do you say that its object is not 
eternal (qadim) also ? 4 

We say : 

The answer to this will follow when we deal with 
the properties of will (irddah). Since we are dealing 
here with power, let us state its properties. 

One of these properties is that it is related to all 
the objects of power (maqdurdt) by which I mean all 
the possible things. Now it is evident that the possible 
things are infinite, and, therefore, that there is no 
end to the objects of power. By saying that the pos- 
sible things are infinite, we mean that the creation 
of contingent things never comes to a point beyond 
which it would be impossible, in reason, for contin- 
gent things to occur. Possibility endures for ever, and 
the power is wide enough to include all that. The 
proof of this assertion, that is, the generality of the 
relation of power [to its objects] is that it has already 
been proven that the creator of the world is one. 
Either He has a particular power vis-a-vis each object 
of power, the latter being infinite, thus establishing 
an infinite progression of such powers — and this is 
absurd because of what has been said earlier about 
the absurdity of infinite progression 9 * — or the power 
should be one so that, despite its oneness, it becomes 
related to all the substances and accidents in their 
multiplicity, because of something common to them 
all. But there is no common element other than pos- 
sibility (imkdn). Therefore, it necessarily follows that 
every possible thing is undoubtedly an object of power, 
and occurs through power. 

Speaking generally, if the substances and accidents 

AlGhazili on Divine Predicates 

proceed from Him. it is impossible that their like do 
D ot>* issues from Him also, because the power to do 
something is power to do its like since 1 - there is no- 
SS prevent multiplicity in the objects o power. 
Therefore, its relation to all movements and all colonrs 
Ti e change! is in one mode. 

[ xSpower] lends itself to the creation of move- 
ment afier movement, perpetually, and hkewise o, 
colour after colour and substance after J^sUnce 
ijawhar), etc. This is what we meant by _say.Bg tha 
His power is related to every possible thing. PossJ* 
litv is not confined to one particular number exclusive 
of others. So it is not possible to point tc , a mov men 
of which it may be said that .t is beyond the possib. 
lity of being related to a power that is also related to 
its like. By necessity we know that what is necessary 
for something is necessary for its like, and from this 
axiom three "'points arise. 13 , tW 

The first point. If someone asks: Do you say »' that 
the contrary to what is known could be an object 
of power? 

This $ matter is controversial, but it would not be so 
if the true nature of the problem were ascertained and 
he linguistic complexity >" removed.The explanation 
is that it has been established that every possible 
thing is an object of power and that an imposs.be 
thing is notLet us, therefore, examine closely whether 
the contrary of what is known is possible (mumhn) 
or impossible {muhal) 1 We shall not know this unt.l 
we know the meanings of impossible (muhal) and pos- 
sible (mumhin) and verify both of them. Otherw.se 
careless investigation may judge the contrary of what 

The Second Pivotal Point concerning the Attributes 

is known at the same time to be impossible (muhal), 
and possible (mutnkin), and to be not impossible. Since 
it is judged to be impossible (muhal), and not impos- 
sible—and two contradictory things cannot be judged 
to be true at the same time — know that there is equi- 
vocation in the terms. This will be shown you by 
what I say, namely, that the world, for example, may 
truly be judged to be necessary (wdjib), impossible 
(muhal), and possible (mumhin). 

As for its being necessary (wdjib), 5 it is called such 
from the following perspective : if the will of the Eter- 
nal a parte ante (qadim) is supposed to exist "• as a 
necessary existence, the object of the will is also cer- 
tainly necessary and not merely probable (jd'iz), since 
the non-existence of the object of will is absurd if at 
the same time the eternal will is verified to exist. 

As for its being impossible (muhal)* it is called 
such if 46 * the will for its creation be supposed not to 
exist. Therefore, its occurrence will be impossible be- 
cause it will lead to the occurrence of a contingent 
thing without a cause, and that is known to be impos- 

As for its being possible (mumkin), 1 it is so called, 
when it is considered by itself, excluding the question 
of the existence of the eternal will or its non-exist- 
ence. Then it will bear the description of possibility 
(imkdn). There are, therefore, three considerations : 

The first is: The positing of the existence of the 
will and its relation to it [i.e. to the mumkin] 11 * as 
a condition. In this respect it is necessary (wdjib). 

The second is: The supposition of the lack of will. 
From this perspective it is impossible (muhal). 

The third is : The omission of attention to the will 


Al-Ghaztil on Divine Predicates 

and the cause. ""^»%£TZ "^ 
£T^*?SS^X m id judgment re- 

this that it » possible » . ittdf, . th£». |we do 
things, bnt it cannot be in tself J™? Theatten . 

a 1 !^^ 1 is contrary to 

Example, we may say that if it is £«* in 
God's knowledge that He ™" ™T t £?L? C te*tL 
Saturday morning, one may f f e ^ s ££ ble 
of life tor Zayd "^J^^tS: ^ 
or impossible. Th V' Ut ? * nsidefation f the divine 
in itself, but ^possible in considerat lon o 

knowledge. The impossible in i self won 
which is unattainable in itself like the bnng g 

attainable because of life itself, but because it is re^ 
ed to an impossibility in something o her han [life] 
itself namely, that very knowledge ( tlm), since 
othrwbe V knowledge would be transformed into 
iglance. and it is absurd that it be so transformed. 
g U is evident, therefore, that it is possible (mum- 
kin) in itself and impossible (muhal) because it is re- 
Sed to an impossibility in something other than 


The Second Pivotal Point concerning the Attributes 

itself. If we say that Zayd's life is an object of power 
at this moment, we mean only that life as such is not 
impossible (muhdt) like the bringing together of the 
colours black and white. And God's power per se does 
not lack relation to the creation of life nor does it 
fall short of creation because of langour, weakness, 
or any reason having to do with power per sc. These 
are two points which cannot be denied. I mean the 
negation of powerlessness in power per $e and the 
affirmation of the possibility of life per se, without 
regard to anything else. If the opponent says that it 
is not an object of power in the sense that its existence 
leads to an impossibility, he is right in this sense, and 
we do not deny that. 22- 

There remains to examine the term [power]. 
The question is whether it is right, in language, to 
use this term or not. It is evident that the right 
thing is to use the term. People say that so and so is 
capable of movement and rest ; if he wills, he moves, 
and if he does not so will, he rests. They also say 
that he has the power at every moment to do both 
these contrary things, though they know that only 
one of them exists in God's knowledge. 23 ' This use of 
the term bears witness to what we have said. Its 
meaning follows necessarily and cannot be disputed. 

The second point. If someone says : You have claim- 
ed that power is generally related to all things pos- 
sible, what do you have to say about the objects of 
the power of animals and all the creatures? Do 
these fail within the power of God or not ? If you say 
they do not, then you contradict } T our premise that 
the relation of power is all encompassing. And if you 
maintain that they fall within God's power, you shall 


Al-Ghazdlion Divine Predicates 

have either to acknowledge a single object of power 
acted upon by 24 * two powerful agents, which is absurd, 
or deny that man and all the animals are powerful, 
which would be an obstinate rejection of necessity 
and a negation to the demands of the Divine Law, 
because it is impossible to demand what cannot be 
done. It is absurd that God would say to His creature : 
You mast do what is within My power and over which 
I have the sole power and you have no power over 
it. 23 * God can never demand from man what God 
knows man is not capable of doing. 

In solving this problem, we say that people have 
taken different positions on the issue. Thejabrites 1 
denied the power of man and were thus forced to deny 
the necessary differences between the tremor and the 
voluntary movement, and consequently had to say 
that the Divine Law made unfulfillable demands. 

The Mu'tazilahs deny the relation of the divine 
power to the acts of man, of animals, of angels, of 
Jinn, and of the devils. 9 The Mu'tazilahs, however, 
claim that all that which issues from man is from the 
creation and origination of man and that God has no 
power to deny or to affirm. There fore, the Mu'tazilahs 
necessarily must affirm two abominable enormities. 
The first is the denial of what the pious forefathers, 
may God be well pleased with them, had agreed 
upon, viz. there is no creator but God and no origina- 
tor other than Him. The second is their attribution of 
origination and creation to the power of the one who 
does not know what he has created. For the move- 
ments which issue from man and the rest of the ani- 
mals, were man to be asked about their number and 
their particularities and their measures, he would not 


The Second Pivotal Point concerning the Attributes 

have a notion about them. For the child crawls from 
the cradle 26 * to suck the breast (of his mother) volun- 
tarily, and the kitten, immediately after birth, crawls 
to the breast of its mother though its eyes are still 
closed. The spiders weave houses in strange forms 
that baffle geometricians 27 * by their circular, parallel 
lines and symmetry of arrangement. We know that 
spiders have no access to that which geometricians 
do not know. 

Bees, too, form their honeycombs in hexagons 
without any square or circular or hectagonal or any 
other form. 28 * This is because the hexagon has a 
characteristic which geometrical proof shows not to 
exist in any other form. The [honeycombs] are con- 
structed according to certain principles, one of them 
being that the most inclusive 29 * and most spacious 
form is the circular which is free from angles made of 
straight lines. The second [principle] is that if the 
circular forms were to be placed in contiguity, aper- 
tures unquestionably would be formed. The third 
principle is that the design closest to the circular in 
its inclusiveness is the form with fewest sides, and 
that is the hexagon. 30 * The fourth [principle] is that 
if all those designs which are close to the circular [in 
shape], like the hectagon, octagon, and the pentagon, 
were to be placed in contiguity and side by side, un- 
employed apertures would be formed and they would 
become uninclusive. The squares can be contiguous 31 " 
but because of the distance between their angles and 
their centres, they are far from being inclusive in the 
way circles are. Since the bees need a form close to 
the circle to encompass their bodies which are almost 
round; and since the space they have is limited, and 


Al-Gkazali on Divine Predicates 

because of their great numbers, they are loathe to 
waste space by creating spaces between their houses' 2 * 
which they cannot use. Since there are no forms that 
are closer to the circular and possessing these charac- 
teristics of contiguity and lack of gaps except the 
hexagon, God has made it practicable to them to 
select the hexagon for the building of their houses. 
Would that I knew, do the bees comprehend these 
subtleties which most sober men are unable to conceive, 
or is it the sole possessor of omnipotence who obliges 
them to attain what they most need ? They are in the 
middle of a course because of the determination of 
God upon it and in it, and they neither know it, nor 
could they disobey it. 

There are in the works of animals of this sort 

wonders which, if some of them were to be mentioned, 

hearts would be awed by the majesty and glory of 

God, Most High. May misery befall deviators from 

the right path of God, those who are deluded by their 

limited power and weak capacities, 33 * who think that 

they share with God in the creation, origination, and 

the bringing into being of the like of these marvels 

and signs. How far and remote the inferior creatures 

are from (sharing with God in creation and origination). 

The Mighty One of the heavens alone possesses 

omnipotence. These are the sort of enormities that 

necessarily follow from the doctrine of the Mu'tazilah. 

Observe, now, the people of the Sutmak, and how they 

were guided to what is right and reared 34 * to the 

golden mean in belief. They maintainthat the doctrine 

of predestination (jabr) is absurd and vain and that 

the doctrine of the origination of bis own acts by the 

creature (ikhtira'a) is a frightful invasion [of God's 


The Second Pivotal Point concerning the Attributes 

sovereignty]. Right, [they say], lies in the affirmation 
of two powers bearing upon one act. 10 The truth lies 
only in the affirmation of two powers operating upon 
the same act and in the doctrine of a single power 
related to two agents. The only thing left to deal with 
is the difficulty of conceiving the coincidence of two 
powers on one act, and this is difficult only if the 
relation of the two powers is the same • but if the two 
powers differ and so does the pattern of their relation, 
then the coincidence of these two relations to a single 
object is not absured as we shall make clear. 
// it is said : 

What makes you affirm one object of power com- 
mon to two agents ? 
We say : 

The decisive proof rests upon the fact that a volun- 
tary movement differs from a tremor or an involuntary 
movement, even if the tremor be supposed to be the 
will of the one who trembles and to be intended by 
him. The difference, therefore, is in power. Further- 
more, the decisive proof is that God's power is related 
to every possible thing. Every contingent thing is 
possible ; and since man's action is a contingent thing, 
therefore, it is possible. If God's power is not related 
to it, then it is impossible. We maintain that in so far 
as a voluntary movement is a contingent possible 
movement similar to the tremor, it is impossible that 
God's power be related to one of them and fall short 
of the other which is similar to it. Nay, another 
absurdity would be required of him, viz. if God were 
to cause man's hand to rest when man wants to move 
it, there would exist either both motion and rest or 
neither, which would lead to the union of motion and 


Al-Ghaz&li on Divine Predicates 

rest or 33 ' to the absence of both. But the absence [of 
both motion and rest], though contradictory, would 
necessitate the negation of the two powers since power 
is what produces the object of power when will is 
realised and the locus is receptive. But if it is thought 
that the object of God's power carries greater weight 
because His power is stronger, this would be absurd 
because the relation of power to one movement is by 
no means preferred to the relation oi another power 
to it, since the result of the two powers is origination 
[of the act]. His power consists in His potency over 
others. His potency over others has no ascendancy in 
this [particular] movement with which we are dealing 
since the destiny of the movement in relation to each 
one of the two powers is that it should be originated 
by it i and the act of origination is equal [in each 
case] so that there is no question of stronger or weaker 
in it so there may be a question of preponderance in 
it. Therefore, the decisive proof affirming two powers 
leads 36 * us to the affirmation of a single object of 
power having two agents. 
If it is said: 

The proof should not lead to an absurdity which 
cannot be understood, and that which you have 
mentioned is unintelligible. 
We say : 

It is our duty to make it understandable [and 
clear]. The clarification is that God's creation of mo- 
tion in a man's hand is intelligible without this mo- 
tion being the object of power of man. Therefore, 
since He creates motion along with a power over it, 
then He has the monopoly over the origination of 
both the power and its object. The conclusion is that 


The Second Pivotal Point concerning the Attributes 

He is unique in origination, that motion is existent, 
and that the mover [i.e. man] is capable of it j and 
because he is capable, his case differs from that of 
the one who trembles involuntarily. All the dubious 
points are, therefore, done away with. The result of 
[this argument] is that the Powerful, who enjoys un- 
limited power, is capable of originating both the 
power and its object. Whereas the names, Creator 
and Originator, are given to Him who produces"- a 
thing by His power, and since both power and its 
object 38 - are products of God's power, He is called 
Creator and Originator. The object of power is not a 
product of man's power even though it is associated 
with it ■ and that is why he is called neither Creator 
nor Originator. It becomes necessary, therefore, that 
this type of relationship 39 * requires a different name. 
The term "kasb" 11 has been applied to the acts of 
men as has been indicated in the Book of God. As 
for the term "act" (fi'L), there was hesitancy in its 
application. In any case there should be no dispute 
over names if the concepts are understood. 

If it is said: 

The aim is the understanding of the concept, and 
what you have mentioned is not understandable be- 
cause it is difficult to understand how the power 
created in man 40 * has no relation to the object of 
power, since power without an object is an absurdity 
like knowledge without an object. And if power is re- 
lated to the object, then the relation of power to its 
object is inconceivable except in terms of effect 
{ta'ihir), bringing into being and the occurrence of the 
object of power through power. The relation between 
power and its object is the relation between the effect 


Al-GhaziH on Divine Predicates 

and the cause, viz. the coming of the object of power 
into being through power. If the object of power does 
not occur through power there would be no relation 
between them' 4111 and power would not be power be- 
cause anything that has no relation is not power, for 
the reason that power is one of those object-taking 

We say . 

Power is relative. Your saying that the relation is 
limited to the occurrence [of the act] through it is 
invalidated by the [analogy of] the relativity of will 
and knowledge. It is also false 12 to say that the rela- 
tion of power is limited only to the occurrence [of the 
object of power] through it since power, according to 
you, endures. 42 * And even if power is supposed to 
exist before the act, [the question will be] is it rela- 
tive or not ? If your answer is negative, this would 
be absurd; and if you say, "yes," then what is meant 
is not the occurrence of the object through it, because 
the object has not yet occurred. Therefore, it becomes 
necessary to establish another kind of relation other 
than the occurrence [of the object] through power, 
because the relation at the time of occurrence is ex- 
pressed by the occurrence through it and the relation 
before that is contrary to this and is altogether a dif- 
ferent kind of relation. Your saying that the relation 
of power has only one pattern is erroneous. Likewise, 
you err regarding the eternal capability (aUqadiriyat- 
ul-qadtmah) because it is related to the world 43 * in 
eternity and before the creation of the world. Our 
saying that it [i.e. power] is relative is true, but our 
saying that the world occurs through it is fallacious 
because [the world] has not yet occurred. If these 


The Second Pivotal Point concerning the Attributes 

were two expressions of one point, then one would be 
true whenever the other is true. 

If it is said: 

The meaning of the relativity of power, before the 
occur ence of the object, is that when the object of 
power occurs, it occurs through power. 

We say : 

This [kind of relation] is not immediate, but rather 
it is an expectation of relation. It would be suitable 
to say that the power is existent while being an at- 
tribute which has no relation, but a relation is expect- 
ed for it when the object of power occurs through it. 
Likewise with respect to the divine power. 44 * This 
position requires an absurdity, namely, that an attri- 
bute which was not one of the object-taking attributes 
has become so and it is absurd. If it is argued that 
the meaning of this is that it is disposed to the occur- 
rence of power through it, we would say that there is 
no meaning to predisposition except the expectation 
of the "occurrence through it" which does not neces- 
sitate an immediate relation. As you would conceive 
of the existence of a power that is related to the 
object of power, though the object does not occur 
through it, by the same token we might conceive of a 
power like that and the object does not occur through 
it but rather through the power of God, Most High. 
So our opinion here does not contradict yours, except 
in our doctrine that the object does occur through 
God's power. But if neither the existence of the power, 
nor its relation to the object of power, necessitates the 
existence of the object of power, then how can its non- 
occurrence 43a through God's power be claimed when 
its existence through God's power has no priority over 


Al-Ghazfili on Divine Predicates 

non-existence with respect to the severance of the 
relationsbipirom the contingent power ; because, if the 
relationship is noUiegated by the non-existence of the 
obiect of power * how then conld [that relationship 
be negated by the existence of the object of power? 
Whether the object of power be conceived as exist ng 
or non-existing. the supposition of an object-taking 
power having no presently existing object isunavoid- 


If it is said: 

A power that does not give occurrence to an object 

of power is on the level oi weakness. 

We sav ' 

If you mean by this statement that the "psycho- 
logical" state which man attains at the time of its 
Re power's] existence, is like the weakness he experi- 
ences at the time of a tremor, then this statement is 
an obstinate rejection of that which is immediately 
evident. If you place it [i.e. power] on the same level 
as weakness because the object of power does not 
occur through it, this is true j but calling it weakness is 
wrong if the reference is to God's power. 13 It would 
be equally absurd, if they, according to their premise 
were to say that power before the act is tantamount 
to weakness because the object does not occur through 
it for the reason that [power] is a perceived psycho- 
logical state, the perception of which differs, in the 
mind, from the perception of weakness. This is similar 
to the former case— there is no difference. 

Speaking generally, we must admit two powers, 
one of which is higher and another which is similar to 
weakness whenever it is related to the higher. You 
have the option between ascribing a power to man 


The Second Pivotal Point concerning the Attributes 

which suggests weakness and ascribing the same thing 
to God. You should have no doubt, however, if you 
are fair, that creatures are more deserving of limita- 
tions and weakness. This is all that this brief summary 
can permit concerning this question. 

The third point. If someone says : How do you claim 
that the relation of power is common to all created 
things when most of the movements 47 * and other 
things in the world are generated, one of them gene- 
rating the other by necessity ? M For example, the 
movement of the hand generates, by necessity, the 
movement of the ring, and the movement of the hand 
in water generates the movement of the water. These 
occurrences are observed and reason also bears them 
out. If the movements of the hand and ring are creat- 
ed by God, it would be conceivable that He may 
create the movement of the hand without the move- 
ment of the water, and that is absurd. The absurdity 
of this position holds true of all generated things with 
all of their ramifications. 

We say : 

What is not understood can neither be rejected 
nor accepted because a doctrine is accepted or reject- 
ed after it becomes intelligible. What is known to us 
by the term "generation" (tawullud) is the coming of 
one body out of the inside of another body in the 
way a foetus comes out of the mother s womb and a 
plant from the earth. But (a similar process) is im- 
possible in the case of accidents because the move- 
ment of the hand has neither an "inside" (jawf) so 
that the movement of the ring may come out of it ; 
nor is it something that contains things so that some 
of what it contains may percolate out of it. What is 


At-Gkazili on Divine Predicates 

the meaning of the generation of the ring's movement 
from that of the band if it [i.e. movement of the 
ring] is not latent in it [i.e. in the movement of the 
handj per se ? This [process of generation] has to be 
made clear > and if [the process] is not understood, 
then your claim that [the process of generation] is 
observed is ignorance and silliness because its occur- 
rence [i.e. of the movement of the water or the ring] 
with it [i.e. with the movement of the hand] is ob- 
served with it and nothing else. 48 * That the one is 
generated from the other, is not perceived. Your 
claim that if [the movement] is created by God/ 9- 
then He would have been able to create the move- 
ment of the hand without that of the ring, and that 
of the hand without that of the water, is sheer non- 
sense comparable to the claim of those who say that 
if the will is not generated from knowledge, 30 * then 
He would have been able to create the will without 
knowledge or knowledge without life. But we say 
that the impossible (muhal) is not an object of God's 
power, that the presence of the conditioned without 
the condition is impossible, and that the condition for 
will is knowledge and the condition for knowledge is 
life ; and likewise, the condition for occupation of a 
place by matter is the vacuity of that place. If God 
should move a [man's] hand, then He would surely 
cause it to occupy a position adjacent to the one 
which it was occupying. If He does not evacuate [the 
place], how could He occupy it with [the hand] ? The 
vacuity of the place is a condition for its occupation 
by the hand. If [the hand] moves and the place is not 
emptied of water by the non* existence of the water 
or its movement [i.e. displacement of the water], 


The Second Pivotal Point concerning the Attributes 

then two bodies would meet in one place which 
is impossible. The vacuity 5U of one place is, there- 
fore, a condition for the other which is the reason 
why they are mutually conditioned and that makes 
it appear as if one is generated from the other, which 
is a mistake. As for consequences (lazimat) which are 
not conditions, we think it admissible that they be 
separate 32 * from the logical conclusion 15 necessary for 
the consequence. Rather their concomitancy is de- 
termined by virtue of following custom, like the 
burning of cotton when it touches fire and the occur- 
rence of coldness in the hand when it touches ice. 
All this continues to happen by the ordinance of 
God ; otherwise, power in itself does not fall short of 
creating coldness in ice and the touching of it by the 
hand along with the creation of heat in the hand in- 
stead of coldness when it [the hand] touches the 
ice. 53 * Therefore, what the opponent sees as "generat- 
ed" falls into two categories. One of them is a con- 
dition which could not be conceived as separate from 
the locical conclusion 54 * and the second is not a 
condition, and hence its separation from the logical 
conclusion 55 * could be conceived if the usual order of 
events is violated. 

// it is said : 

You did not prove the negation of generation but 
only denied that it is understood. It is, however, 
understood because we mean by it neither the bringing 
forth of a movement from another by its [i.e. move- 
ment's] coming out of [a movement's] inside, nor the 
generation of coldness from ice by the coming forth 
or coldness out of ice, and its transference, or its com- 
ing forth out of the coldness itself. Rather, we mean 


Al'Gkazali on Divine Predicates 

by it the existence of an existent as a consequence to 
an existent and its being existent and originated by 
it [i e. by the existent]. Therefore, we call the origi- 
nated thing generated, and that by which generation 
occurs, is called generator [i.e. that which generates]. 
This naming is understandable; what, then, points 
to its negation ? 

We say : 

If you accept 5 ** that [i.e. the above argument], 
then what indicates the falsity [of your position] is 
the same as has indicated the absurdity of the con- 
tingent power's being a creator. Therefore, if we think 
it impossible to maintain that an object of power is 
originated by a contingent power, why should we not 
consider as impossible the occurrence [of an act] by 
that which is not power ?"• Therefore, its impossibility 
[Le. of generation] is due to the relativity of power 
being common; and its exclusion [i.e. exclusion of 
generation] from the sphere of power negates this 
common aspect of its [i.e. power's] relativity. But 
this position is absurd and, furthermore, necessitates 
impotence and mutual hindrance 58 * as has been stated 


Yes, the Mu'tazilah who hold [the doctrine] of 
generation (tawallui) have fallen into innumerable 
contradictions in their exposition of generation ; such 
as their claim that inquiry generates knowledge, while 
its recollection [i e. recollection of knowledge] does 
not generate it [i.e. knowledge] , 59 * etc. There is no 
need for verbosity in what is not necessary. You have 
understood from the summation of this [i.e. the above 
discussion] that all contingent things, their substances 
and accidents, which occur in the essences of living 


The Second Pivotal Point concerning the Attributes 

beings and inanimates, occur through God's power 
and that He is unique in their creation, and not one 
of the created things occurs through another [i.e. none 
by contingent power or tawallud] but rather all occur 
through God's power. What we have wanted to clarify 
is the affirmation of the attribute of power in God, its 
[i.e. power's] general properties, and the aspects and 
consequences associated with it. 


1. Ghazali in this chapter formulates a concept of divine power which is 
essentially atomistic. This view is based on the basic Ash'arite principle 
which is God's absolute power, a power which cannot be limited by any 
consideration. The principle which underscores this concept of God's 
power is that it is related to every possible thing. Only the impossible 
cannot be an object of power (translation, pp. 4-5). This is so because the 
question whether God Who composed the parts of a body is powerful enough 
to separate them until they are reduced to the atom or not must be asked. 
If the answer is negative, this would be ascribing to God powerlessness 
[fa" jit). But if the answer is affirmative, then the opponent would admit the 
concept of the indivisible atom (translation, p. 8). Ghazali denies that 
there can be infinite division in actu and hence infinite regress It seems by 
this denial Ghazali is trying to refute Nazzim's claim that a body is divi- 
sible ad infinitum (Intisdr, pp. 32-3). But Nazzam is not speaking there 
about a division in actu ', rather he talks about an intellectual division 
(ibid., p. 33). 

Ghazali restates once more here his and the Ash'arites* doctrine of 
causality, a doctrine which denies all natural laws and man's freedom in 
acts. All natural phenomena are here dismissed as occurring through mere 
custon (ford al-'&dah), (translation, p. 19). Ghazali says that every contin- 
gent thiog and every act or movement is divisible into separate parts and 
that power creates movement after movement, and substance after sub- 
stance perpetually (translation, p. 4). This, Ghazali says, is true of bodies 
and not accidents, for accidents do not endure for two moments of time. 
He goes on to prove this through the example of the ring on the hand 
when man moves the hand. Here Ghazali says, there is no causal nexus 
between the movement of the hand and that of the ring. He concludes by 
saying that man does not Influence these movements, but he certainly can 
relate to himself an act which God has created. This power of man is what 


Al-Ghaz&H on Divine Predicates 

he called kasb (translation, pp. 13, 16-7). 

Al-Ash'ari in the M aqaUt (pp. 377-8), sums up for as the various 
Mn'tazilah positions concerning the concept of man's power as follows : 

Some say : 

The Maker of the world is able to make His creatures "create" sub- 
stances, colours, flavours, and all other kinds of acts. This is the claim of 
the extremists among the Rifidis. 

Others say : 

God cannot be described as enabling His creatures to "do" substances, 
but He is powerful to make them "do" all accidents such as life, death, 
knowledge, and the power to do all other kinds of accidents. 

This is the position of al-Silihi. 

Some say : 

The Maker of the world is powerful to eoable His creatures to [do] 
colours, flavours, coldness, moisture, and dryness. He is more powerful 
than they in so doing. As for power on life and death, it is not permissibls 
that He enables them to do any of these. 

This is the claim of Bishr ibn al-Mu'tamar. 
Others say : 

There i* no accident which is not permissible for God to empower [His 
creatures] to do its like. The only accident, according to this group, is the 
movement. As for colours, flavours, coldness, and sound, they have denied 
that God could empower His creatures to do them because, to them, these 
are substances, and God can only empower His creatures to [do] move- 

This is the statement of an-Nazzim. 

Some say : 

It is possible that God may empower His creatures to do movements 
and rest, sound, and pain and all that which they know its "howness 
tkayfiry**) As for accidents whose "howness" thev do not know sue* as 
colours, flavours, life, death, powerlessness. and dower, it is not permis- 
sible to describe God as able to empower [His creatures] to do any of them. 

This is the position of Abu-1-Hudhayl. 

2. See Chapter I'Al-Iqtisad Fil-I'tiqdd, Metlmdand Importance" pp t-2. 

3. To the Mn'tazUah the problem was whether attributes, conceived as 
real incorporeal beings distinct from God's essence, existed in God or not. 
Their interpretation of attributes affirmed of God as operative or Negative 
attributes was used by them as a means of saying that none of the terms 
predicated of God in the Quran or in the common speech of men are to be 
taken to signify the existence of real attributes in God. Ghazili takes issue 
with them. for. logically with the denial of real attributes, all the terms 
affirmed of God become predicates which are identical with the subject. 
Such a tautological approach is not acceptable to Ghazili because this 
would be tantamount to saying God is God See Wolfson : («) "Avicenna, 
Alghazali. and Averroes," Homenaje a MiOas-Vallicrosa. Barcelona, 1956, 
Vol. II, pp. 545-71 : (6) "Philosophical Implications of the Problem of 
Divine Attributes in the Kalim." Journal of tke American Oriental Society. 
Vol. 79. pp. 73-80 : (c) "Maimonides on Negative Attributes," Louis Ginsberg 
Jubilee Volumes. v\>. 411-46. 


The Second Pivotal Point concerning the Attributes 

4. Ghazili 's contention Is that an act could never be eternal. If it were 
eternal, it would cease to be an act, for an act is something which is started 
after it was not. 

5. Ibn Sini in 1 shir it (pp. 447-9-1) defines the necessary (wijib) as that 
which by its essence is necessary if we do not take into consideration any- 
thing other than its very essence. If it is necessary by itself, then this would 
be the "truth". But also if the condition for the existence of the necessary 
is supposed to exist, then the necessary must exist. 

6. The impossible (mukdt) or. as Ibn Sini calls it, the mumtani' is that 
which either by its essence is impossible or because its cause is non- 
existent (ibid., pp 447-9-1). 

7. The possible [muntktn) is that which cannot exist by itself, for its 
existence has no priority over its non-existence. Its existence or non- 
existence is only due to the existence of something or its non-existence. By 
"something" Ibn Sini means the cause. Tiisi, commenting on this says that 
Ibn Sini means that the possible can only exist if it has a cause different 
from it. The stress here is that there should be a determining factor 
(murajjih) (ibid., pp. 448-10-1). 

8. The Jabrites, headed by Jahm ibn Safwia, maintain that man's act 
is not his own, that God alone is the creator of all acts. In this respect, 
they claim there is no difference between an involuntary act such as the 
tremor of the hand or the act of falling from a high place, on the one hand, 
or such acts which man imagines are his own acts such as walking, speak- 
ing or motion. Man is completely compelled {majbQr) because he is devoid 
of any power or will. He is like a leaf in stormy weather. God alone creates 
for him his acts and runs them through him The Jabrites, therefore, attri- 
bute power to man in a metaphorical sense. No man h3S power in reality 
[kaqiqak) {Maqiiat, p. 279 ; Mituridi. Skark ot-Fiqh al-Akbat, pp 9-10). 

9. Some Mu'tazilah deny that God has a power over the acts of men or 
animals. They attribute such acts to man's free will (ikhiiyir). We have 
discussed this in various parts above (Intis&r, pp. 53-4 ; Maq&lat, pp. 377- 
8. 563. 566). 

10. The Ash'arites tried to steer a middle course between the Jabrites 
who deny man any power and the Mu'tazilah who attribute to man a power 
over his acts. The Ash'arites thus affirmed two kinds of acts : (a) involun- 
tary acts in the face of which man is completely powerless, which shows 
that they are definitely created for him and over which he exercises no 
will, and (6) voluntary acts over which man has power preceded by will. 
Such power is what makes man acquire (yaktasib) his acts. Kasb or acquisi- 
tion is this association between man's power and the act of God which means 
that if maa wills an act, God would at that very moment create for him a 
power to do it and thus man acquires the power {Maq&I&t, p. 542). 

U. This term kasb which we have discussed in n. 10 above is derived 
from the Qur'ao where it appears in many places to mean what man ac- 
quires of sins and good deeds. According to S. Pines [Madkhab at-dharrak, 
p. 31). al-Ash'aii formulated it from Us generic form as expounded 


Al-Ghazfili on Divine Predicates 

by Dirar and al-Najjar {M aqdldt. pp. 383. 408* 566). Al-Ash'ari's own 
dtfinition is as follows : "The truth as I believe is that the meaning of al- 
iktisib is that a thing occurs through a contingent power and thus it 
becomes acquisition (hash) to him through whose power it occurs" (ibid., 


12. This is a reference to those philosophers who take an anti-atomist 
position and maintain that power endures. For Ghazili it lasts one 
moment. This is in accordance with his doctrine of continuous creation. 

13. Ghazili is for treating the attributes as positive attributes of action, 
but the possibility of his treatment of these attributes as negation may be 
expended, although* soch statement could not serve as a groundwork for a 
negative aspect. We feel, however, its documentation is within the realm 
of possibility, hard as it may prove. 

14. The Mu'tazilah differed on the concept, generation (tawallud). Some 
of them maintained that it is the act which occurs through me and alights 
in other than me. Others said, it is the act which / determine its cause and 
thus it becomes beyond my ability to abstain from it. I may do it wUhtn 
myself or in other than myself. Some said: it is the third act which succeeds 
mv will such as pain which succeeds a blow or motion [dhthSb) which 
succeeds a push {daf'ah). Al-l.kifi said : any act which occurs through a 
mistake or without being intended or willed is generated (mutawalltd). 
while any act which occurs only through intention and every part of it 
requires renewed intention {qosd). falls outside the boundary of generation, 
and is to be included within the boundary of the direct act (mubishir) 
(Maqdldt, pp. 408 ff.). Consider also Ma'mar's views that all generated 
things are the acts of bodies naturally (ibid., pp. 564 ff., 405). 

15. See note 11, above.* 


Chapter I J 


{•1LM) ' 

We assert that God knows all existent and ^non- 
existent objects of knowledge^^^^ 
ed into eternal and contingent things The eternal 
things are His essence and His attributes. Whoever 

and attributes It follows, by necessary, thai He know s 
His own essence and attributes if it be a «. tb ^ 
He knows other than Himself. It.skno wn tha t He 
knows other than Himself because that to which the 
name "other" is applied is His masterly ^creation .uA 
perfectly organised acts wh.ch pm * *« »« ™ ™ ^ 
of the Maker just as they pant to His po^r. « >™ 
have seen earlier. Whoever sees arranged line.. «mng 
in harmonious form from a scribe, and then douDis 
his ri r^ribe's] knowing the craft of writing would 
fa suiy'in his doubt. Therefore, it has been established 
that He knows His essence and others than Himself. 

If it is ashed: ,. .,, 

Do His objects of knowledge have a limit ? 

■S'STi things existing at the present - 
JL. the possible things in the ^ "JJ*™*- 
He knows'" the possibles which are non-existent, ana 
SneftTHe shall actually create them or no and 
These are infinite.- He, therefore, knows what is in- 
finite Nay if we are to multiply a thing >n various 


Al-Ghai&li on Divine Predicates 

ratios and measurements, this would be infinite, but 
God knows them all. We say that, for example, the 
double of two is four, and the double of four is eight, 
and the double of eight is sixteen ; and in the same 
manner if we multiply the multiple of two and the 
multiple of the multiple ad infinitum, a man would 
know of their [i.e. these mathematical calculations] 
degrees only what his mind is capable of knowing and 
his life will end while the infinity of the multiplications 
continues. Since the knowledge of the multiples of the 
multiples of two (and it is but a simple number) are 
unaccountable— and it is the same with every number 
like it— then what of the other ratios and quantities ? 
This knowledge, and its relation to infinite objects of 
knowledge, is but one as will be clarified later along 
with the rest of the attributes. 2 


1. The Mu'tarilah did not deny the attribute ol knowledge but made 
equal to the essence. Al-'Allaf said that God is knowing throngh a know- 
ledge which is His essence. Affirming knowledge to God meant negating its 
contrary which in effect is that knowledge is attribute of negation, i.e. it 
negates ignorance to God. An-NaMim. however, declared that it is a. 
attribute of negatioc and that all existent things are known to God. i e are 
unveiled to God's knowledge, and it is not an attribute superadded to toe 
essence. He also added that it is an eternal attribute which does oot 
change, for. if it changes. God would be a locus for contingent things and 
what is not free from contingent things is also a contingent thing. To tt»e 
objection that since object, of knowledge are subject to mutation and 
change a. can be perceived by the senses, why does not kmWp chnnge 
also » They answer that change is true ol the knowledge of Man because it 
is . kaowledge which is acquired by the senses aod the sense, only sense 
what is changeable. But God know, a thing before it occur, and afler ,t «. 
and know, when it is annihilated. Change in time lead, to change in man . 
knowledge but the co-cept of time cannot be applied tp the divine know- 
ledge because the revelat io » of time to God's knowledge is one. All thing. 


The Second Attribute : Knowledge ('Ilm) 

are known to God eternally without their chanjeability effecting ch«ge. 
T^Hi. knowledge or making Him kadUk {M^Slat PP- ^T). 

* n~i tr»n« everything, whether universal or particular, but the 

Forth. M.-t«tt.h .t «.tbo»gbttha« « God k.o« th. p«t.c.l«. that 
... Sad^d-M. +TMlO*m i« hi. com»«tary •» tb. 'M* <* >f "—f- 
knowledge when it is connected with tnem. *.„«v«„w 

«!.««. impo»ibl. or po-ibl.. contl.g«.t or «t«o.>. .Ota*, or faf. 
particular or onl«nal. Tbi» i« «o becwwe : ..... ., w .._vH.. 

That ,»«.«.... b«o« ho... i. th. ---£*• obl ' C, ■ °' ta, °" te,Ue 

^TSS^^ZSZlSt^!^^ i. JJ-- 

si sxzu. .. — .*>«*• -r*?u: ssrsr ptsr 

ctaog. io th. .-eotlal koo«l«dg. ? Taftlrtal'. »•«. .. that God . too. 
ledg. of changeable thing, i. ol two aqxcU : 

4 A knowledge which is ^not bound Iby ^^JJ^^^^l^y 
every one of them bound *y.thetim e Df the »' £« fen their nonexistence. 

and bound by «^*»E^E Jj^ SaW Another knowledge 
Knowledge, however, is eternal and «» "^ " JJ ^^n change-that 
is bound bT time and th.t }■ G^« JJ <£* a g c finite ,„ mam .ccord. 
thing, are eitlier existent or disappearing ana iD» p^tgntia like the 

ing fo the nniteoes. •« ^^JSJSE-S thei oeit£r oecessitate. a 
eternally changeable thing.. But the «""■""" teaX io God's eewnce. 

change fn the attribute «X££&to£E^^*» *° d iU ^ 
Rather it necessitates a change m tne reimwi. w 

nection with its objects (p 84>. ^ |tar vie _ 
Al-Dewwini in Skork 0--A 1 a>Ual--Adadiyyak supports a similar view. 
For him. God know, all object, of knowledge . 

wr.T£ Da— M M— « •*«« ■**••* »» PW">"* , '»* " ho d " T 
that God know, thopatticalara. 

being a consequence of the sen is a perfection (p 109) 

XsSykh Abual-Barakit al-Baghdadi awerts »t (pp. UO-1). 


Al-Ghaz&li on Divine Predicates 

To difcuss the position of the philosophers. al-Dawwani says that there 
are four points to be considered : 

111 Things either have no form and are non-changeable; 

(2) or they are changeable bat do not have forms ; 

«3) or they have forms but do not change ; 

(4) or they have forms and do change (p. 111). 

As for (1), God knows tham whether they are universal* or particulars- 
How could it be said that He does not know the particulars while most of 
these philosophers are agreed that He is knowing in virtue of His essence 
and knows the intelligences. 

As for (3) they ace like the heavenly bodies whose measures and forms 
are permanent and free from change. And they are. in their view, not 
known to God individually because the perception of bodily things is done 
only by a bodily instrument. 

As for (2), they are like the forms and the accidents and the rational 
souls. They are not objects of knowledge for God. not because they require 
• bodily instrument but because, since they are changeable, their change 
necessitates change in knowledge. 

As for (4). they are like the existent and corruptible bodies. They are 
impossible to be objects of knowledge for the two reasons that they are 
changeable and subject to corruption. 

Later philosophers replaced "having form" by "being material ana 
maintained that He does not have changeable-material-particulars. iney 
considered them as they are. i.e. inasmuch as they are material— J-e. 
attached to matter and its accidents and inasmuch as they are changeawe 
and occupying units of time, in the past, the present and the future, una 
because the perception of material particulars since their attachment to 
matter makes them objects of sensing and imagination. Sensing and imagi- 
nation are only done by a particularised bodily instrument and also the 
perception of the changeable particulars which occupy units of time, their 
change necessitates change in knowledge. 

Al-Dawwini takes these points posed by the philosophers as they are 
related by Tusi and refutes them one by one. 

The first argument, he contends, was not necessary because God is free 
from matter and its attachments and hence there is no need to deny sens- 
ing and imagination— which are properties of matter— to God. The second 
argument, he declares, is hardly necessary because God's rationalising is 
not through a faculty like our human rationalising faculty. Thirdly, al- 
Dawwini argues that he does not accept the supposition that if God s 
knowledge is through rationalising that would not refute a supposition of 
isktirtk because God's knowledge of His essence and of the intsllects is in a 
particular way. Fourthly, because to contend that "being universal" and 
"being particular" is one of the absolute qualities of knowledge is im- 


Chapter III 


We assert that He, Most High, is living. This is 
known by necessity, and none of those who have 
acknowledged that He is Powerful and Knowing deny 
it. That a powerful knower is living, is necessary, be- 
cause we only mean by the living what is conscious 
of its self, 2 knows its self and others than itself. How 
could the knower of all objects of knowledge and the 
powerful over all objects of power not be living? 
This is evident, and the investigation into the attri- 
bute of life should not occupy us any further. 


t. The attribute of life raised no controversy of significance among the 
Mu'tazilah or the philosophers except in the way this attribute is to be 
predicated of God which falls within the general problem of predication. 
Even Ibn Rnshd devoted to it a few lines in ManShij al-Adillah, pp. 161-2, 
where he says that the existence of the attribute of life is apparent from 
the attribute of knowledge because one of the conditions of knowledge is 
life. He further says that what the Mutakallims have said in this respect is 

2. That a living is what is conscious of itself is a concept that opens 
many possibilities except that Ghazali chose not to elaborate on it here 
for it hints to the question of the ego and existence as is formulated by 
Ibn Sina in Kitib au-Nafs of at-Najdt. 


Chapter IV 

We assert that God wills His acts; and the proof 
[of this assertion] is that an act issuing from Him is 
characterised by sorts of possibilities which cannot 
be distinguished from each other except by a prepon- 
derant. 2 His essence is not sufficient for preponderance 
because the relation of the essence to the opposite 
things is the same. What is it, then, that causes one of 
two opposite things to occur at a specific time and 
not the other? Likewise, power is not enough in this 
case as the relation of power to two opposite things is 
one and the same thing. Moreover, knowledge {'Um) 
is not sufficient — contrary to al-Ka'bl, 3 who found 
knowledge sufficient without will — because knowledge 
succeeds the object of knowledge and is related to it 
as it is; and [knowledge] neither affects [the object] 
nor changes it. If a thing is possible in itself, and is 
co-extensive with another possible thing which is op- 
posite to it, then knowledge is related to it per se. We 
should not make one of the two possible things pre- 
ponderant over the other : rather we should intellect* 3 * 
two possible things and intellect their co-extensiveness. 

God, Most High, knows that the existence of the 
world at the time of its actual existence was possible, 
and that its existence either after that time or before 
it was equally possible because these possibilities are 
co-extensive. Knowledge, therefore, should be related 
to [the world] in all these stages of possibilities. If the 


The Fourth Attribute : Will 

attribute of will requires its [the world's] occurrence 
at a specific time, knowledge would be related to the 
determination of its [i.e. the world's] existence at that 
time because of the relation of will to it [i.e. to the 
actualisation of the world]. In this case, will 64 * would 
be the cause of determination, and knowledge ('Urn) 
would be both related to it and subordinate to it with- 
out affecting it. If it were permissible that knowledge 
is sufficient without the will, it should be sufficient 
without the power. Rather, that [i.e. knowledge] 
would be sufficient for the existence of our acts and we 
would not need the will since one side becomes pre- 
ponderant because of the relation of God's knowledge 
to it and that is absurd. 

If it is said: 

This [sort of reasoning] would be turned against 
you concerning will itself because just as the eternal 
power is not related to one of two opposite things to 
the exclusion of the other, the eternal will is likewise 
not determined for one of two opposite things to the 
exclusion of the other, for, its specification of either 
of the two opposite things should be through a specifier 
which leads to infinite regress, since it may be said 
that the essence [alone] is not sufficient for causing 
fa thing] to come into being for the reason that if it 
[i.e. a thing] comes into existence from the essence, it 
would exist concomitantly with the essence, not later. 
There should be a power • but power is not enough 
because if a thing comes into existence not because of 
the existence of power, it would not have been speci- 
fied by this [particular] time 3S * since the relation of 
time — before and after the coming of a thing into 
being — to the possibility of the relation of power to it 


Al-Ghazall on Divine Predicates 

[i.e. to time] is in one mode. 

What is it, then, that specifies [this particular] 
time? [If the answer is], it needs will [to be so speci- 
fied], it would be said that will [alone] is not enough 
because the eternal will, like the eternal power, is 
general in its relation [to its objects] and its relation 
to all time is one just as its relation to two opposite 
things is also one. Motion, for instance, occurs instead^ 
of rest because will is related to motion rather than to 


Tt may [further] be asked: 

Is it possible that [will] be related to rest ? 

If the answer is negative, this would be absurd. If 
the answer is affirmative, then motion and rest are 
co-extensive in their relation to the eternal will. What, 
then, makes it a necessity for the eternal will to pre- 
ponderate motion rather than rest ? It [i.e. will] would 
require a preponderant [in this case], and it would, 
therefore, be necessary to ask about the determining 
principle of the determining principle, ad infinitum. 

We say that this is a question which has puzzled 
the minds of people of various schools, and none but 
the Sunnis were divinely guided to the truth. Scholars 
are divided into four groups over it. 

Some say that the world exists through the espence 
of God, Most High, and that it [i.e. essence] absolute- 
ly has no added attribute. 6 ** Since the essence (dhdt) 
is eternal, they maintain, therefore, the world is eter- 
nal; and the relation of the world to it [i.e. to the 
dhat] is like the relation of the effect to the cause, the 
light to the sun, and the shadow to a man. These are 
the philosophers. 4 Another group maintains that the 
world is contingent (hddtth) but came into existence 


The Fourth Attribute : Will 

at the time when it came into existence, neither be- 
fore that nor after it, because of a contingent will 
that had occurred to it, but not a locus (mahal), 5 [i.e. 
not in God's essence], thus requiring the coming of the 
world into being. These are the Mu'tazilah. 

Still a third group maintain that it [the world] 
came into existence through a contingent will in His 
essence. This is the group maintaining that He [i.e. 
God] is a locus for contingent things. 8 The fourth 
group assert that the world comes into existence at 
the time when the eternal will is connected with its 
occurrence [at that time] without the origination of a 
will, and without a change in the attribute of the Eter- 
nal. 7 Consider the various groups, and compare the 
position of each to the other. The path of none of these 
groups is free of insoluble doubt except that of the 
Sunnis whose doubtful aspect is easy to resolve. 

As for the philosophers, they have asserted the 
eternity of the world/which is absurd; because it is 
impossible for the act to be eternal, since the^ meaning 
of its being an act is that there was a time when 
it was not [existent], then it became [existent]. If it 
were eternally existent with God, how could it be an 
act? The assumption of infinite revolutions would 
necessarily follow from this, according to what we have 
stated before, and that is absurd from [many] aspects. 
Moreover, with their falling in this difficulty, they 
cannot rid themselves of the basic question, which is: 
Why is will connected with the coming into being of 
the contingent thing at a specific time rather than be- 
fore or after that time, in spite of the co-extensiveness 
of the relations of the. Various units of time to the 
will ? Even if they solve the problem of the specifica- 



Al-Ghazali on Divine Predicates 

tion of time, 67 " they would not solve the problem of 
the peculiarity of the attributes, because the world is 
specified by a measure and a specific arrangement 
though the contrary of all such ,8 [measures and ar- 
rangements] is rationally possible to conceive. The 
eternal essence is not related to some possible things 
to the exclusion of others. The most grievous error 
which this position requires of them, for which they 
have no excuse and from which they have absolutely 
no escape, consists of two points which we have men- 
tioned in the book "The Destruction of the Philoso- 
phers" (Tahajut airFaldsifah)* 

One of these points is that some of the revolutions 
of the spheres are easterly, that is from the east to the 
west, and the others are westerly, that is from where 
the sun sets to where it rises. 68 * The reversal [of these 
easterly and westerly revolutions] is co-extensive in 
possibility because directions are equal for revolutions. 
How, then, was it necessary for the eternal essence or 
for the heavenly spheres, which they consider eternal, 
specifically to single out a direction and not the one 
opposite to it which is co-extensive with it from all 
aspects ? To this they have no answer. 

The second is that the highest sphere, which is the 
ninth sphere, in their view is the one that moves all 
the heavens once during the day and night [every 
twenty-four hours] by compulsion, revolving on two 
poles, northern and southern. The poles, so to speak, 
are the two opposite points on the surface of the 
sphere which are fixed at the time of the revolution 
of the sphere on itself [i e. on its axis]. The equator 
(al-manfiqah) is a huge circle around the centre of the 
sphere that is the same distance from the two poles. 


The Fourth Attribute : Will 

We say that the mass of the highest sphere is 
symmetrical and similar. Any point on it [i.e. on its 
surface] could be considered as a pole. What made it 
necessary, then, that two points were singled out [to 
be poles] from amongst the mass of points which they 
assert to be infinite ? There must be an attribute, dif- 
ferent from the essence, whose function is to distin- 
guish a thing from its like. That [different attribute] 
is nothing but will. We have given these two ob- 
jections their due of careful examination in the book 
[Tahafut] "Destruction of the Philosophers". 

As for the Mu'tazilah, they have rushed blindly 
into two abominable and false subjects. The first is 
that the Beneficent, Most High, wills through a con- 
tingent will which is not a locus [i e. not in His 
essence]. If will does not subsist in Him, then the 
assertion of those who say that He wills through it 69 * 
is improper use of language, like their saying that He 
wills through a will subsisting in something other 
than Himself. The second is : Why does will occur at 
this specific time ? If it occurs through another will, 
then the examination of this other will becomes 
necessary, and this will lead to infinite regress. If it 
occurs without a will, then the world should occur at 
this specific time without will which is absurd. 70 * But 
this is absurd because the need of a contingent thing 
to will is because of its possibility, not because of its 
being a body, or a heaven, 71 * an earth, or a man j 
and all contingent things are equal in this [need for 
will]. Furthermore, the [Mu'tazilah] cannot rid them- 
selves of the problem because they would be asked 
why wtll (trddah) occurs at this specific time and why 
willing motion occurs rather than willing rest. Since 


Al-Ghazili on Divine Predicates 

in their view, there occurs for every contingent thing 
a contingent will which is related to that contingent 
thing, why, then.-dees not the contingent will come 
into being related to the opposite of that contingent 

thing ? 

As for those who take the position of asserting the 
origination of will in His essence, they have avoided 
one of the two problems, viz. His being a wilier 
through a will in something other than His essence, 
but they have added another problem, viz. His being 
a Locus for contingent things, which necessarily 
makes Him originated (hadith). Therefore, the rest of 
the problem remains for them, and they have not rid 
themselves of the question. 

As for the People of the Truth, they have maintain- 
ed that contingent things originate through an eternal 
will which is connected with them [i.e. with contin- 
gent things], and which distinguishes them from their 
opposites which are similar to them. The question 
of those who want to know why it [i.e. will] is related 
to them [i.e. with the contingent things] while their 
opposites are equally possible is a false question be- 
cause "wiH" is nothing other than ah attribute, the 
function of which is to distinguish one thing from its 
equal. The question of those who ask why will dis- 
tinguishes one thing from another is like the ques- 
tion of those who want to know why knowledge 
('Urn) makes necessary the disclosure of the objects of 
knowledge {ma'lum). It would be said that "know- 
ledge" is meaningless if it does not disclose the objects 
of knowledge. Therefore, the question of those who 
want to know why jt (*ilm) renders necessary the un- 
covering [of the object of knowledge] is like their 


The Fourth Attribute : Will 

asking why is "knowledge" knowledge in itself ; why 
is the "possible" possible j and why is the "necessary" 
necessary? All this is absurd. 72 * "Knowledge" is 
knowledge in itself and so is the "possible," the 
"necessary," and the rest of the essence. Likewise 
is will, whose essence is to distinguish one thing from 
its like. Therefore, the statement of those who want 
to know why it [will] distinguishes a particular thing 
from its like is like their saying why is the "will" will 
and the "power" power. That is absurd. Each [of the 
four groups] is obliged to confirm the existence of an 
attribute, the function of which is to distinguish one 
thing from another equally possible thing. This [attri- 
bute] is nothing other than will. Therefore, the best 
of the [four] sects in doctrine and the most rightly 
guided are those who affirm this attribute [i.e. will] 
and who also do not think of it as a contingent [will] 
but maintain that it is eternal [and] connected with 
contingent things at a specific time. Therefore, the 
occurrence [of contingent things] at that time is be- 
cause of this [i.e. because of the relation of the eter- 
nal will to the contingent things at a specific time]. 
No sect can do without this [thesis], and by it 
the necessity to continue discussing this question is 

Now that our thesis concerning the principle of 
will has been put in order, know that according to us 
[will] is related to all contingent things inasmuch as 
it has been clarified that every contingent thing is 
created through [God's] power and that everything 
created through power needs will to dispose power to 
the object of power and to particularise it [i e the 
object] with it. Since every object of power is willed, 


Al-Ghaxtllt on Divine Predicates 

and since every contingent thing is an object of power, 
therefore, every contingent thing is willed [i.e. is an 
object oi will]. Evil, infidelity, and sins are contingent 
things and, therefore, are objects of will. What God 
wishes is, and what He does not wish is not. This is 
the doctrine of the venerable fathers and the creed of 
all the Sunnls, and its proofs have been furnished. 

As for the Mu'tazilah, they say that all sins and 
all evil originate without God's will ; nay, He hates 
them [i.e. He hates evil and sins]. 9 But it is known that 
most of what happens in the world is sin. Therefore, 
what He hates is more numerous than what He wills. 
According to their [i.e. Mu'tazilah's] assertion, He is 
nearer to impotence and inability. God is loftier than 
the claims of the unjust. 

If it is asked: 

How does He command what He does not will, and 
how does He will something and then forbid [practis- 
ing] it I and how does He will immoral acts, sins, trans- 
gression and what is ugly when the wilier of what is 
ugly (qabi h) is foolish ? 

We say: 

If we examine the reality of "command" and de- 
monstrate clearly that it is different from "will," and 
examine the [nature] of the ugly and the good, and 
clearly demonstrate that they are due to their agree- 
ment with human predilections, or to their disagree- 
ment with [human predilections]— God is far above 
predilections— these problems will be overcome." This 
will be dealt within its proper place if God so wishes 


The Fourth Attribute : Will 

1 We have discussed the attribute of will at length in our Introduction. 
Here* we may and that Ibn Rastad rejects completely Ghazali's contention 
that will is eternal and dismisses the notion that that in which contingent 
things subsist must be contingent cannot be substantiated or defended 
scientifically (Mandhij al-AdVlah. p. 162). 

2. The concept of murajjik is Ghazalis most sustained argument which 
proves the existence of an eternal will. 

3. Abul-Qasim ibn Muhammad al-Ka'bi was a student of al-Khayyit. 

He maintained that : „ 

God's will is not an attribute subsisting in His essence, that He is 
neither willing in virtue of Himself nor is His will generated in a locus or 
no locus. When He is called willing, it means that He is knowing. P *«*™J 
and is neither forced to do an act nor is He hateful of it. II * *•«"** 
He wills His acts, this means that He creates them according to His know- 
ledge lAl-Milal ma at-Nihat. p. 53). m \ 

4. The reference is to Ibn Sini and al-Farabl who maintained that the 

"TmsTthe school oi Abul-Hudhayl al-'AUaf. the head of the Mu-ta- 
zilah. He is said to be the first Mu'tazili to speak of a will not in a locus 
[if Adah Id ti mahal) [Al-MUal ma aJ-Ni**/. pp. 34-5). 

{ 6 The KartUite. are the followers of Abu 'Abdallah Muhammad .bn 
Karrim Shahrastani consider, him among the «ttrib.tbU *«£j£ 
affirmed attributes to God except that his extreme position led him to rank 
among the Corporealists and the likeners The Karramites ■"£«* 
subsistence of most contingent thing, in God'.essence They «•*■*•""• 
that what occurs in His essence occurs through His power, that what 
occurs and is different from His essence, occurs through contingent creation 
(at-iMdth). and by the latter they mean the "act of creation \al-%}**) ana 
the "act of annihilation" {al-i'ddm) which occur in His essence through 
His power of speech and will. By generated (mukdath) they mean all that 
which is different from His essence such a. substances and accidents.^ Th» 
way they differentiate between -creation" and "creature and between the 
"act of creation" and the existent and the creator (mujtd) (Atrial mo at 

Nikal, pp. 79-81). " ^ ■ . . . ,. m 

7. As is consistent with the Ash'arite doctrine of the eternity of the 
attributes, they maintain that the eternal will doe. not change as would 
the contingent thing . 

8. See Tahdful al-Faldsifak, I. p. 58. 

9. See note, on Wilt. 


Chapter V 


We assert that the Maker of the world is Hearing 
and Seeing. We substantiate this assertion by the 
Divine Revelation arid by reason. As for the Divine 
Revelation, numerous verses from the Qur'Sn vindicate 
our assertion. The Most High says : "He is the Hearer 
and the See-er." And like the saying of Abraham, 
peace be upon him, "Why do you worship what does 
not hear, see, or suffice you a thing? " We know that 
the statement cannot be turned against him [i.e. Abra- 
ham] with respect to his own object of worship and 
that he worshipped a hearer and a see-er; otherwise, 
he [i.e. Abraham] would have been the partner of 

If it is said; 

By [hearing and seeing] is meant knowledge. 

We say ; 

The terms set by the Divine Revelation should be 
transferred away from their [seemingly] intelligible 
and immediately understood meaning if it were impos- 
sible to suppose them to mean what has been set forth ; 
but there is no impossibility in His being Hearer and 
See-er • nay, He should be so. It is senseless to deny 
arbitrarily what the people of the consensus (ijma') 2 
have understood from the Qur'Sn. 

// it is said ; 

The aspect of the impossibility [of His being Hear- 
er and See-er] is that if His hearing and seeing were 


The Fifth and Sixth Attribute concerning Seeing and Bearing 

contingent things. He would be a locus for contingent 
things, and that is absurd. On the other hand, if 
they [i.e. hearing and seeing] were eternal, then how 
would He hear a voice that is non-existent, and how 
would He see the world in eternity, when the world 
is yet non-existent, for the non-existent cannot be 

We say: 

This question would issue from a Mu'tazili or a 
philosopher. As for the Mu'tazili, his refutation is easy 
because he has admitted that the Most High knows 
all contingent things. Therefore, we say that God 
knows, now, that the world has been existent before 
this [moment]; how, then, does He know in pre-eternity 
that it was existent when it has not yet existed? 
If it is possible to establish an attribute in pre- 
eternity, which becomes [at the time of] the existence 
of the world a knowledge that [the world] exists be- 
fore [the existence of the world] that it will be, and 
after [the world has come into existence], that it exist- 
ed without this attribute undergoing any change, this 
attribute is expressed by the term "knowledge" {"Urn) 
and the state of being knowing ('alimiyyah). 1 ^ This 
you can properly apply for hearing — and the state of 
being bearer, seeing — and the state of being see-er. 

If [the question] issues from a philosopher who 
denies that God knows the particular 74 * contingent 
things which fall within time, past, present, or future, 5 
then our course [in answering the philosopher] would 
be to bring the discussion of "knowledge" ( € ilm) and 
to establish [against his argument] the permissibility 
of an eternal knowledge which is connected with con- 
tingent things, as we shall mention later. If that can 


Al-Ghaz&ll on Divine Predicates 

be established concerning knowledge, we would apply 
it analogically to hearing and seeing. As for the ration- 
al course, we would say that it is known that the 
creator is more perfect than the created ; and that it is 
also known that the seeing is more perfect than the 
one who does not see, and the one who hears is more 
perfect than the one who does not hear. Therefore, it 
would be impossible to affirm the attribute of perfec- 
tion to the creature and not to the Creator. There are 
two premises which necessitate admitting the sound- 
ness of our assertion. Concerning which [of these two 
premises] is the dispute ? 

If it is said: 

The dispute is in your saying that it is necessary 
that a creator be more perfect than the created. 

We would say: 

This should be admitted by the [force] of the [pro- 
visions] of the Divine Revelation— as well as ration- 
ally. The Muslim community (ummah) and intellectuals 
are all agreed upon it. This [kind] of question does not 
issue from a believer. He whose intellect can accept 
[the existence of] a powerful agent who is able to 
create what is higher and nobler than he is, would be 
stripping himself of the human instinct [which rejects 
such absurdity], and his tongue would be uttering 
what his heart would shrink from accepting if it [i.e. 
his heart] is to understand what [i.e. the tongue] has 
uttered. For this [reason] we do not see intelligent 
persons who believe in such a doctrine. 

If it is said: 

The dispute concerns the second premise, viz. your 
saying that a see-er is more perfect and that the facul- 
ties of hearing and sight are perfection. 


The Fifth and Sixth Attributes concerning Seeing and Hearing 

We would say: 

This too is self-evident because knowledge is per- 
fection, and the hearing and seeing are secondary 
perfections 4 for knowledge. We shall demonstrate 
that they [hearing and seeing] are a kind of comple- 
mentary perfection for knowledge and im? ination. 
Whoever knows something which he does not see, and 
then sees it, gains further knowledge of that thing and 
perfection. How could it be said that such [perfection] 
occurs to the creature and not to the creator ; or how 
could it be said that it is not perfection? If it is not 
perfection, then it is defect, or it is neither defect nor 
perfection. 75 * Any of these classifications is absurd. 
Therefore, it is clear that the right [opinion] is what 
we have said. 

If it is said: 

You are bound to admit this regarding the percep- 
tion acquired by the faculties of smell, taste, and 
touch because their absence is imperfection, and their 
presence is perfection in perception, because the per- 
fection of the knowledge of one who knows the flavour 
is not like the perfection of the knowledge of one who 
perceives by the faculty of smelling and likewise by 
the faculty of taste. It is one thing to know about an 
object of taste and quite another to perceive it by the 
faculty of taste. 

The answer is that authoritative scholars have un- 
equivocally affirmed various kinds of perceptions' 
together with hearing, sight, and knowledge, 7 * which 
are perfection in perception to the exclusion of the 
causes that are customarily connected with them, 
like touching and contact because these are improper 
to God. They also permit that the faculty of sight may 


Al-Ghaz&li on Divine Predicates 

perceive without the object of sight being opposite to 
it. 6 In the application of this analogy is the answer to 
this question, and nothing prevents its use. But since 
the Divine Revelation provided nothing except the 
terms: "Hearing," "Sight/* and "Knowledge," it 
would be impossible to predicate others. Imperfection 
in perception could in no way be possible for God. 

If His said: 

This would lead to the affirmation of pleasure and 
pain for God. A numb man who does not feel the pain 
of beating is imperfect, and an impotent 77 * who does 
not delight in sexual intercourse is imperfect Likewise, 
the loss of appetite is an imperfection ; so it is proper 
to attribute appetite to God. 

We say : 

These matters point to origination. They are in 
themselves, if examined, imperfections and require 
matters which in turq require 78 * temporal origination. 
Pain is imperfection and requires a cause which is a 
"blow". And the "blow" is a contact occurring be- 
tween bodies. Pleasure, if investigated, [would be found 
to be] due to the disappearance of pain or the attain- 
ment of what is needed and desired. Appetite and 
need are imperfections ; therefore, what depends upon 
imperfection is imperfection. Appetite means the 
quest for what is suitable, and there is no need except 
when what is sought is lacking and no pleasure except 
when what is non-existent is obtained. All that which 
is possible for God to make existent is existent. He 
does not lack anything so that He may become 
desirous by seeking it and pleased by obtaining it. 
How could these matters, then, be conceived about 


The Fifth and Sixth Attributes concerning Seeing and Hearing 

If it is said: 

Failing to suffer from and feel the beating are 
imperfections for the benumbed j that its [i.e. pain of 
beating's] perception is perfection — that the lack of 
appetite in one's stomach is imperfection and its pre- 
sence a perfection, 19 * viz. it is perfection in relation 
to its opposite which destroys it; becoming, thus, 
perfection in relation to destruction because imperfec- 
tion is better than destruction. It is, therefore, not 
perfection in itself contrary to knowledge and the 
above-mentioned perceptions. This much you should 


1. These two attributes Ghazali considers as perfections oi the living, 
and the knowing. 

2. By AM al-Ijmd'. Ghazali, merely means those learned men whose 
wisdom and excellence in the sciences of religion are attested to by the 
community of Muslims. 

3. See note 2 to Chapter II above. 

4. According to Ibn Sina, the soul is the first entelechy (perfection or 
actuality) of a natural body possessed of organs. This is the most compre- 
hensive definition of the soul, which, when added to the differentiae, would 
yield the definition of the species, so to speak. "Every species of the soul 
includes the lower soul and transcends." The qualification of perfection by 
"first" suggests, as F. Rahman points out, a distinction between two kinds 
of actuality corresponding to knowledge or thinking {Al-Naj*t t Maq&l&h ft 
an-Nafs, pp. 157-8 ; Avicenna's Psychology, p. 72). 

5. Probably Ghazali means mystical perception which is considered by 
Sufis as the light kind of perception. 

6. Ghazali is pointing here to the btatific vision which Ghazali has dis- 
cussed at length from a Neoplatonic standpoint in Iqtisad. pp. 60-73. He 
rejects the idea that vision could only be acquired if the object of vision 
were in a direction. That discussion could be summed as follows : 

Those who claim that for a thing to be visible implies its being in a 
direction build their argument on *he basis of the fact that they have never 
seen anything except in a direction. Thus they pass judgments on every- 
thing on the basis of existent beings only. But if it were possible for us to 


Al-Ghaxili on Divine Predicates 

pass judgment on everything on the basis of concrete examples, the 
anthropomorphists (Mujassimmh) would be right in their claim that 
God is a body, because God is an Actor, and they have not seen any actor 
except in a body. Hence those who hold that for a thing to be visible must 
be in a direction are no better than the anthropomorphists. 

Further, most of the Mu'tazilah accept that God sees Himself and the 
world, though He is not in a direction either in relation to Himself or in 
relation to the world. Once this is accepted, their whole argument about 
the impossibility of God's Vision on the assumption that for a thing to be 
visible it must be in a direction falls. For if God can see Himself and the 
world without being in a direction. He can also be seen without His being 
in a direction. Pp. 62-63. 

Ghazill concludes his discussion by saying that vision is a kind of 
perception and that only those who are pure from material attachments 
could see God. Excepting Ghazali's idea that God can be seen by the eye. 
not different from the Mn'taxilah who say that God shall be seen by the 


Chapter VI 


We assert, as is the consensus of all Muslims, that 
the creator of the world is speaking. You should know 
that those wishing to affirm divine speech through 10 * 
the premise that reason judges permissible that the 
creatures oscillate between command and prohibition, 
and through the premise that any attribute which is 
permissible for the creatures depends upon an attribute 
necessary for the creator, are grossly mistaken be- 
cause the proponents of such views would be told : if 
you mean the permissibility of their [i.e. the creatures] 
being commanded by their [fellow-creatures] of whom 
speech is conceived, this would be accepted ; but if 
you think of such permissibility as being common 
to both the creatures and the creator, then you 
would be begging the question, and this could not be 
accepted. If one attempts to affirm the divine speech 
on [the authority of] the consensus and/or on [the 
strength] of the Tradition of the Messenger, ODe would 
wrong oneself because the consensus leans on the 
Tradition of the Messenger * and if one denies the Bene- 
factor's being speaking, one necessarily denies the 
concept of messengership since being a messenger 
means transmitting the speech of the sender. If it is 
inconceivable that the one who asserts that he is a 
sender speaks, how could a messenger be conceived ? 
We do not, for example, listen to one who tells of 


Al-GhazM on Divine Predicates 

being the messenger of the earth or the mountain to 
us, because we believe in the impossibility of a speech 
or a message from either the mountain or the earth. 
To God alone is the highest ideal. 

Whoever believes in the absurdity of attributing 
speech to God could not possibly believe in .a messen- 
ger, because whoever deems the divine speech a lie 
would necessarily deny its transmission. The message 
is an expression of the transmission of the speech, and 
the messenger is an expression of the transmitter. It 
seems that a third premise is the correct course which 
we have followed in affirming [to God], "hearing" 
and "sight/' viz. that speech is [an attribute] of the 
"living". It is said either that [speech] is perfection 
or imperfection— or that it is neither imperfection nor 
perfection— but since it is absurd to say that it is 
not imperfection and/or neither perfection nor imper- 
fection, it would be necessarily established that it is 
perfection. Therefore, every perfection which is an at- 
tribute of man is a fortiori a necessary existence for the 
Creator 31 ' as we have pointed out above. 
If it is said: 

The speech which you have made the source of 
your speculation is the speech of the creatures. 82 * By 
that is either meant the sounds and letters, the power 
to produce them, or a third meaning distinct from 
these two. If what is meant is the sounds and letters, 
they are contingent things and some contingent things 
are perfections to us but could not be imagined to sub- 
sist in God's essence. If [speech] subsists in something 
other than Him, He would not be speaking through 
it, but rather the locus in which [speech] subsists 
would be the speaker. 83 * If what is meant is the power 


The Seventh Attribute concerning Speech 

to produce sounds, then it is perfection • but a speaker 
is not a speaker because of his power to produce 
sounds only, but because of his creation of speech in 
himself. God is able to create sounds and unto Him 
alone is the perfection of power. God does not speak 
through [sound] unless He creates that sound in Him- 
self which is absurd because, by creating sound in 
Himself, He would be a locus for contingent things. 
Therefore, it is impossible that He be speaking. If by 
speech a third meaning is meant, this would be unin- 
telligible and the affirmation of what is unintelligible 
is absurd. 2 
We say : 

This classification is accurate and the question in 
all its dichotomies is acknowledged -except for the 
denial of the third category [i.e. inner speech]. 3 We 
admit the possibility of the subsistence of sounds in 
His essence and the absurdity of His being speaking 
in this manner. We say, however, that man is called 
speaking from two perspectives: 

(a) Through sound and letters, i.e. vowels and 

(6) Through Jnner speech (hadithu-n-nafs) which 
is neither sound nor letter and which is per- 
fection not impossible to God and does not 
imply contingency. 
We affirm to God the inner speech only. There is do 
way to deny man the inner speech which is something 
distinct from the power to speak and distinct from 
sound. A person would say: "Last night I made up a 
speech in mind (fi nafsi)." It is also said : in the mind 
(nafs) of so and so is a speech and he wants to utter 
it. The poet says : 


Al-Ghaz&lt on Divine Predicates 

Be not bewildered by 

The writing of one created 

'Till his calligraphy be 

With the speech equated. 

For, speech is in the mind 

But by the tongue articulated. 4 
What poets utter indicates a self-evident concept in 
whose perception all men are equal and which could 
not be denied. 
1/ U is said: 
. This interpretation of inner speech is acknowledg- 
ed. It, however, neither lies outside the knowledge 
and perception M * of things nor is it absolutely a genus. 
What scholars call inner speech (kalamu-n-nafs) and 
inner whispering (hadithu-n-nafs)*** is the skill of 
combining terms, phrases and the composition of in- 
telligible concepts which are known in specific order. 
There is nothing in the mind (qalb) except intelligible 
concepts which is the "knowledge," and heard terms 
which are known through hearing and these, too, are 
the knowledge of what terms are set for, 84 * and to 
that is added the composition of concepts and the 
arrangement of terms in specific order, such process 
is called cogitation (fikr ) and the power producing such 
process is termed the cogitative faculty (quwwah 
mufakkirah) If you establish, however for the mind 
something other than this very cogitation (fikr ) which 
is the arrangement of terms, concepts and their com- 
position ; and if you establish something other than the 
cogitative faculty (quwwah mufakkirah), which is the 
power [that arranges terms, concepts and their com- 
position] ; and if you establish something other than 
the knowledge of separate and combined concepts, 


The Seventh Attribute concerning Speech 

and if you establish something other than the 
knowledge of terms which are composed of individual 
and combined letters, then you will be establishing an 
atrocious matter unknown to us To clarify this we 
maintain that speech is either command or prohibi- 
tion, either informative or interrogative statements. 

As for the informative, it is a statement indicative 
of a knowledge in the mind of one who declares it. He 
who knows something and knows the term designat- 
ed 17 * to indicate that thing like "beating" (darb), for 
example, a concept which can be perceived by the 
senses and is a term composed of the letters {dad, ra' 
and ba') t designated by the Arabs to indicate a sen- 
sible concept which is another knowledge— he has 
the ability to acquire these sounds with his tongue, 
the desire to indicate and the desire to acquire the 
term, then when his statement "beating" (darb) is 
uttered, he would not be lacking a thing further than 
these matters [to inform of the concept "beating"]. 
His saying (darb) would be complete; it would be a 
declarative statement (khabar) and speech. Anything 
you judge aside from this would be easy to refute. 

As for the interrogative (istikhbar), it is an indica- 
tion of a demand to know. Command, however, is an 
indication of a demand to do what is commanded. 
Prohibition and all other parts of speech are analogi- 
cal 27 to this. No speech could conceivably fall out of 
these forms of speech. Of these forms of speech, how- 
ever, some, like sound, are impossible to God, and 
others are existent to God like will, knowledge, and 
power. Anything else is unknown to us and is unintel- 

But the speech which we mean is a concept 


Al-Ghaz&li on Divine Predicates 

distinct from these forms of speech. So let us deal 
with it within the scope of one of the parts of speech, 
viz, the command (amr) so that our explanation will 
not be unduly long. 
We say : 

A master's command to his slave : "Stand up 1 " 
{qutn) $ is a term indicating a meaning. The meaning 
indicated is in itself a speech. But this you have not 
taken into consideration and,therefore,there is no need 
for hair-splitting. To refer this command, viz. "Stand 
up," to a desire to command 88 ' or to the desire to 
indicate an existing state of affairs, could be wrongly 
suggested. But it is absurd to suggest that it is a desire 
to indicate a state of aff iirs because the indication 
requires an object of indication which is neither an 
indication nor a desire to indicate. It is equally absurd 
to suggest that it is the desire of one who commands. 89 ' 
One may command without wishing compliance < nay, 
one may [in fact] dread compliance as in the case of 
one who makes excuses to a sultan intent upon killing 
him as punishment for his beating his slave, by saying 
that he only beat his slave because of his disobedience, 
the evidence being that he would command the slave 
in the presence of the [sultan] and the slave would 
disobey. Now, if he becomes litigious and commands 
the slave in the presence of the [sultan] by saying : 
"Stand up— I am giving you a decisive order for which 
you have no excuse 90 * to disobey," and the slave 
would not stand, then at that moment he is unques- 
tionably commanding [the slave] to stand up, but he 
is [also] unquestionably not wishing the slave to obey 
the order. Any fair-minded person would judge that 
the command which subsisted in the master's mind 


The Seventh Attribute concerning Speeth 

and which was indicated by the imperative is speech, 
yet altogether distinct from the desire to stand. 

If it is said: 

This man is not commanding in reality but rather 
feigning a command. 

We would say : 

This is false for two reasons : 

(a) If he were not really commanding, his apology 
would not have been easily accepted by the* [sultan] 
and he would be told : command could not be 
imagined from you at that moment because yours is 
a demand for compliance and it is inconceivable that 
at that moment you would want compliance which 
would be the cause for your destruction. Further- 
more, how could you desire to prove something by 
disobedience 91 ' to your command when you are in- 
capable of really commanding it because you are in- 
capable of wanting that in which lies your destruc- 
tion and in the slave's compliance lies your destruc- 
tion. Now, there is no doubt that he is capable of 
using such a proof and that his proof stands and 
earns the acceptance of his apology and that the 
proof 92 ' is established by disobedience to the com- 
mand. So if the command could not 9 " be conceived 
despite the master's hatred of the slave's compliance, 
we could absolutely not conceive his use of it. This 
is decisively convincing to those who ponder it. 

(6) If the [master] tells this incident to the mufti 
and swears a final divorce that after the sultan's re- 
primand, he commanded the slave to stand up, in 
the sultan's presence, and that the slave did not obey 
him, every Muslim would decide that his divorce had 
not taken place. The official expounder of Islamic 


Al-GhazdH on Divine Predicates 

law [tnufti] <* nnot say : " X kD ° W that !t b * m P° t S9ibl , e 
thlt at that moment you would want the slave's 

compliance, wlich would cause your destruction. Com- 
mand implies* he wish of compliance and if you have 
commanded, 9 " then you must have wished compli- 
ance " This f .uftVs view Would be invalid by the con- 
sensus. TV presence of a concept which is indicated 
by the expression and which is distinct from what the 
opponents have enumerated 954 above has, therefore, 
been unmasked. We call this [concept] speech, and it 
is a genus different from all types of knowledge, wills, 
and beliefs ; the affirmation of which is not impossible 
ta God i nay, it should be attributed to Him because 
it is a kind of speech 96 *— the eternal speech we mean. 
As for letters (hur&f), they are contingent and 
are indications of speech. The indication is neither 
the object indicated nor does it possess its attributes 
even if its indication is essential like the world which 
is a contingent thing that indicates an eternal Maker. 
How, therefore, is it unlikely that contingent letters 
would indicate an eternal attribute although this is 
an indication by convention ? 

Since the question of inner speech is a subtle one, 
it defies the understanding of those who are intellec- 
tually weak, who affirm only letters and sounds. 
Concerning this doctrine [various] questions and ob- 
jections will be directed against them, some of which 
we will mention so that a method of refuting other 
problems may be delineated. 

The first [objection] concerns the question of those 
who ask how did Moses (Masa) hear God's speech. 

They ask : 

Did he hear sounds and letters ? If your answer is 


The Seventh Attribute concerning Speech 

affirmative, then according to you he [Moses] dM I not 
hear God's speech because God's speech is neithe by 
Sers nor sound; and if he did not nearletters^ or 
sound, how could he hear what was nether letter 
nor sound ? 

bute subs.st.ng in Gods *°*™^ fl did hc 

r^^SS^S^l-- iron, one 
iho dots n P ottnae«tand. what is -quired of the 
Nation ''how". One should understand what is re- 
Sed by ' W' and by what it can be answered » 
E to know the absurdity of the quest.on. 

Hearing is a kind of perception and the question : 
Hearing » * JT h question of one who 

IZ Ho, do^u perceive sweetness of sugar by 

£ orgaTof 7aste P ? Tbere are only two ways of de- 

cisively answering this question : ^ 

(1) By giving su 8 a ' ".^^^d sweetness. 

^LTsi tZ" ■ ' ba" Reived it just as you 
Then we say xo mm satisfac- 

SUC f 2 ) a The Loud way applies when the first way is 
imposible eitherlor the lack of sugar or for the »ac k °f 

ihe^questioner's having tasted sugar. The ef ore J 
tell him • "We perceive its flavour just as you pe£ 
SL the sweetness of honey." This «■■« «■«£ 
correct from one aspect and wrong from another The 
correct aspect is that it is a definition of something 


Al'Gkazili on Divine Predicates 

similar to the one in question in one aspect, viz. the 
principle of sweetness, though it is not similar to it in all 
aspects because the flavour of honey is different from 
the flavour of sugar. So if it is comparable to it in one 
aspect, viz. the aspect of sweetness, this kind of com- 
parison is all that is possible in this case. But if the 
questioner had never tasted the sweetness of anything, 
it would be impossible to answer him and to make 
him understand what he had asked about. He would 
be like the impotent man who asks about the pleasure 
of sexual intercourse though he has never experienced 
it. It would be difficult to make such a man under- 
stand except by Ukening sexual intercourse to the 
pleasure of eating which would be a false analogy be- 
cause the pleasure of intercourse and the state which 
the participant attains is not at all similar to the state 
perceived by the eater except inasmuch as the general 
concept, pleasure, has included it. If he had never 
been pleased, no answer would be possible. Like- 
wise as for anyone who asks : How did Moses hear 
God's speech?, he could be satisfactorily answered 
only by our making him hear God's eternal speech 
and that is impossible for us because that is one of 
the privileges of Moses, and we are incapable of mak- 
ing the [questioner] hear it, or by comparing it to 
something which he [the questioner] has heard No- 
thing that he hears is like God's speech because all the 
things heard with which the [questioner] is familiar 
are sounds, and sounds are not similar to what is not 
a sound, hence the difficulty in making him under- 
stand. Moreover, if a deaf man asks : How do you hear 
sounds [though he never heard a sound] ?, we would 
not be able to answer him, because if we say : As you 


The Seventh Attribute concerning Speech 

perceive visible things for it [hearing] is a perception 
in the ear, like the perception of sight in the eye. This 
would be erroneous, because the perception of sounds 
is not like the seeing of colours. This example points 
to the absurdity of the question. Moreover, if someone 
asks* How do you see the Lord of Lords in the here- 
after? to answer him would be absurd because he is 
asking'about the mode (kayfiyyah) of that which does 
not have a mode. What the questioner means by 
"how is it" is "like which thing of what we know is 
it?" Since what he is asking about is not similar to 
anything he knows, the answer would be impossible 
and this does not point to the non-existence of God s 
essence. Likewise, the impossibility of such an answer 
does not indicate the speechlessness of God. Rather, 
it should be believed that His speech is an eternal 
attribute, having no likeness, just as His essence (dhat) 
is an eternal essence which has no likeness. And just 
as we see His essence in a manner different and unlike 
our seeing bodies and accidents. His speech is heard 
in a manner different and unlike the hearing of letters 
and sounds. 

The second objection. // * « ^ked : Is God s 
speech located in copies of the Qur'an {majdhtf) or 
not? 5 If it [God's speech] is located there, how does 
the eternal reside in the contingent? If you say no, 
this would be contrary to the consensus, because the 
respect of a copy of the Qur'an (mushaf) is so much 
demanded by consensus that the ritually impure is 
forbidden from touching it. This is so only because 
God's speech is in it [in the mushaf]. 

We. therefore, say that God's speech is written in 
the books (masahif), preserved in the hearts, and read 


Al-Ghaz&li on Divine Predicates 

by tongues. But as for paper, ink, writing, consonants, 
and vowels, they are all contingent things because 
they are bodies and accidents in bodies and all that 
is contingent. And if we say it is written in the 
books (masahif), (I mean the attributed the Eternal, 
Glory to Him), this does not require that the eternal 
is located in the book (mushaf), just as, if we say that 
"fire" is written in the book, it would not necessarily 
follow that the essence of ,fire is located in it because 
if it [i.e. essence of fire] resides in [the book], the book 
(mushaf) would be burnt, and he who utters [the word] 
fire, if the essence of fire is located in his tongue, his 
tongue would be burnt. The fire (tiar) is a hot body 
having a verbal sign which consists of the articulated 
sounds that produce the [letters] nun, alif, and ra 
Therefore, the hot [body] that burns is the essence 
of the thing indicated and not the sign itself. Like- 
wise, the eternal speech that subsists in God's essence 
is the thing indicated and not the sign. Letters are 
signs and signs have a sanctity because the Divine 
Law (shar') made them sacrosanct Because of this 
[sanctity], respect of the Book (mushaf) is required, 
because it embodies a pointer to God's [eternal] attri- 

The third objection is their demand : Is the Qur'Sn 
Godf's speech or not ? If you say no v you violate the 
consensus, and if you say yes, then what is it except 
the letters and sounds? It is acknowledged that a 
reader reads by letters and sounds. 

We say ; 

There are three expressions involved here: "read- 
ing" (qira'ah), "what is read" (maqru') and "Qur'an". 
As for what is read (maqru') , it is God's speech, i.e. 


The Seventh Attribute concerning Speech 

His eternal attribute which subsists it .His lessen^ 
As for the "reading" feW«*]. *» on the tongue as 
an expression of the act of the reader [qarp who began 
it. after he had abstained from it, and [to «J tWi 
thing is] contingent means only that it has been 
beeun after it was not. If the opponent does not un- 
Sstand this [meaning] from the [term] contingent 

(kadith), then let us abandon ^V'™ Vu^ act 
Lnd created, and say that reading (qtrSah) is an act 
begun by the reader (,«><') af ter he was not doing rt 
and it is sensible [i.e. perceived by the senses] As 
for the [term] -Quf-dn," it could be applied as mean- 
ing what is read (tnaqru'). If this is what is meant by 
it. then it is eternal, not created. This is what he 
pious forefathers meant by their saying that the 
Qur'Sn is God's speech, not a created th u>( J '-mean- 
fog what is read (.naqru') by the tongues And jf what 
is meant by it is the reading itself (?.r« 'ah) wh ch is 
the act of the reader (qari'), then the act of the 
reader does not precede the reader's existence, and 
what does not precede the existence of the contingent 
Ihadith) is contingent. And in sum, he who says that 
he [the reader] has originated by choice, consisting 
of sound and its articulation through letters and be- 
fore which he was silent, is eternal, should not be 
oartner to discussion or obligated but, rather, it is 
suitable that he be known as a poor fool who does 
not know what he says, neither knowing the meaning 
of letters nor the meaning of contingent. If he knew 
these things, he would know that if he is himself 
created, then what issues from him too is created 
and he would know that the eternal is not transfer- 
able to a contingent essence— so let us not dwell on 


Al-Ghaziii on Divine Predicates 

what is self-evident. He who says "in the name of 
God" [bismi-lldhi), if the sin is not after the [ba'] t 
it would not be Qur'an, but rather it would be an 
error and if the sin is after something else and fol- 
lowing it, how could it be eternal ? We mean by the 
eternal a parte ante that which never follows any- 
thing else. 

The fourth objection is their saying that the com- 
munity (ummah) of Islam is unanimously agreed that 
the Qur'an is a miracle of the Messenger of God 
(God bless him and give him peace), and that it is 
God's speech, and that it consists of chapters (suwar, 
sing.surah) and verses [ay at), having periods and 
beginnings. This being so, how [they ask] could the 
eternal have endings and beginnings, and how could 
it be divided into chapters and verses, and how could 
the eternal be a miracle of the Messenger when a 
miracle is an unusual act, contrary to nature, and 
when every act is created? How then could the 
speech of God, Most High, be eternal ? 
We say : 

Do you deny that the term [Qur'an] is commonly 
applied both to the act of reading (qird'ah) and to 
what is read (maqrW) ? If you acknowledge this fact 
[i.e. the equivocality of the expression "Qur'an"], 
then [we may observe] that all in all that the Muslims 
have mentioned 97 * in describing the Qur'an as eternal, 
like their saying that the Qur'an is God's uncreated 
speech, they meant what is read {al-maqru'). And in 
all that with which they have described it which the 
idea of eternity does not permit, as its being chapters 
(suwar) and verses having endings and beginnings, 
they meant expressions that the point to the eternal 


The Seventh Attribute concerning Speech 

attribute, which is the act of reading (qird'ah). If the 
term is considered equivocal, the contradiction would 
no longer exist. The consensus is that only God is 
eternal; and God, Most High, says, "till it returns 
like an aged (qadtm) palm- bough" But we say that 
the term "eternal" is common [i e. equivocal, mush- 
tarak] to two concepts. If it is affirmed from one 
aspect, it would not be impossible to negate it from 
another; the same applies to the term "Qur'an 9 *. 
This is an answer to all the contradictory designa- 
tions which they reiterate even if they deny its being 
equivocal. We definitely know its [i.e. the term 
Qur'dn's] application is to the meaning of what is real 
{al-maqru) and that it is clearly shown by the pro- 
nouncements of the pious forefathers, that the Qur'&n, 
God's speech, is not created despite their knowledge 
that they, their voices, their readings, and their acts, 
are all created. As for its application to the act of 
reading (al-qiraah) 98 * the poet has said : 

They have sacrificed the hoary-headed man on 
whose forehead prostration marked a sign, 7 One who 
divides the night between praising and Qur'dning, 
meaning "reading". The messenger of God, God bless 
him and give him peace, said, "God has not granted 
permission to a prophet as he has permitted me the 
prolonging and modulating sweetly [my voice] in 
reciting the Qur'an." Prolonging and modulating 
sweetly the voice is done in the act of reading. All the 
pious forefathers have maintained that the Qur'an, 
God's speech, is not created, that it is a miracle which 
is the act of God. Since they have known that the 
eternal a parte ante could not be miraculous, it is 
evident that it is an equivocal term. 8 He who does not 


Al'GhaiSli on Divine Predicates 

understand the equivocality of the term would imagine 
contradictions in these applications. 

The fifth objection is the assertion that it is acknow- 
ledged that nothing but sounds are heard now [i.e. in 
this world], and God's speech is heard now accordiog 
to the consensus of the community and the evidence 
of God's saying, "And if any of the idolaters seeks of 
thee protection, grant him protection till he hears the 
words of God." 

We say: 

If the sound which the idolater (mushrik) hears at 
the time he seeks protection is God's eternal speech 
which subsists in His essence, what merit does Moses, 
peace be upon him, have over the idolaters, in his' 
being called the one who spoke with God, since they 
hear as he had heard. 

No 99 * answer to this objection can be conceived 
except by saying that what Moses, peace be upon him, 
heard is an eternal attribute subsisting in God's es^ 
sence, and what the idolater hears is sounds pointing 
to that attribute. What makes the above description 
quite clear is the idea of equivocality, either in the 
expression "speech," in which case it is calling the 
signs by the name of that to which the signs point, 
"the speech" here is definitely the inner speech [of 
God], As for the terms, because they point to it, they 
are called "speech" just as they are called knowledge 
('*/*»). If you say: I have heard so and so's know- 
ledge, you hear only his speech that points to his 
knowledge, or [the equivocality lies in the expression] 
"what is heard," because known by hearing something 
other than it could be called an object of hearing just 
as you say I have heard the speech of the prince (amir) 


The Seventh Attribute concerning Speech 

from the tongue of his messenger when it is acknow- 
ledged that the prince's speech does not subsist in the 
tongue of his messenger, but what is heard is the utter- 
ance of the messenger that is indicative to the prince's 
speech. This is what we have wanted to mention for 
the clarification of the doctrine of the Sunnites [Ahl 
as-Sunnah] concerning the inner speech which is 
thought to be enigmatic. We shall mention the re- 
mainder of the properties of speech when we deal 
with the properties of the attributes in the second 
section if God, Most High, so wishes. 


1. The Ma'tazilah considered God's speech as similar to ordinary speech 
which is composed from vowels and consonants. By speech they meant the 
act by which a speaker expresses the concepts in bis mind. Since the Qur'an 
is composed of sounds and letters and these being contingent, the Qor'an 
then mast be contingent. Tbe Qar'in, they argued, is not eternal because 
it is not an attribute of God, but rather an act of His. God creates speech 
in the "tablet" or in the angel Gabriel or in the prophets. The Mu'tazilah 
pointed to many verses of the Qar'in which would support their claim 
that the Qar'au is created. There are in the Qur'an such verses that tell of 
the Qar'in being sent down in chapters and parts and this certainly is a 
characteristic of a contingent thing {Al-Milal. pp. 35, 68, 46). 

Al-Ash'ari and the Ash'arites in general believed in two kinds of speech, 
one of them being mental and they call it inner speech (kaldmu-n-n*f$) 
which is an essential attribute of God. The other kind of speech is the 
ordinary speech composed of words and letters The former is God's speech 
and is eternal while the second is contingent. Al-Ash'ari maintains that 
God's speech is one and it is "command and prohibition, informative or 
interrogative statements, promise or warning". These various aspects of 
the divine speech are due to considerations related to tbe speech and not 
to a multiplicity in the speech or expressions. The terms which descend to 
the prophets are indications of the eternal speech while the indications 
themselves are contingent. According to al-Ash ari. a speaker is tbe one in 
whom speech subsists. Tbe Mu'tazilah, however, consider the speaker the 
one who "does' ' the speech. They, however, consider the expression, speech, 


M-Ghaz&tt on Divine Predicates 

ither metaphorically or equivocally (Al-Milal w al-Nihal. pp. 67-8 V 
/Jthoach al-Mataridl does adhere to the Ash'arite concept of the divine 
speech, he criticises them when they say that the physical book conld be 
borot because God's speech is an attribute and the attribute cannot be 
separated from the one described He says this Is "idiocy" because what is 
annihilated is known through God's knowledge. He then asks: Do yon see 
that the attribute of knowledge wonld be annihilated because what is 
annihilated is known ? Likewise, he adds, is God's speech ; it cannot be 
described as being subject to annihilation because of what is written in the 
copies [masikif) of the Qur'an We do not say, he adds, that the speech is 
not alighting' in the copies of the Quran so that it may be described as 
subject to annihilation {Sharh al-Fiqh al-Akbar, p. 2G). 

2. See note 1 above. 

3. KaUtmu-n-nafs is al-Aah'ari's unique concept through which he and 
later Ash'arites tried to explain what they considered to be God's eternal 
speech. See note 1 above. 

4. Fr>m a poem by the poet al-Akhtal. 

5. See note 1 above. 

6. See note 1 above. 

7. The reference is to the murderers of 'Uthman, the third Khalifah. 

8. In Islamic philosophical works, "restatements of Aristotle's distinc- 
tion between 'equivocal mushtarak and 'univocal* mutawatik, terms usually 
contain another type of terms which stand midway between these two. It 
is called 'ambiguous' or 'amphibolous' mustarikah terms" (Wolfsoo, "The 
Amphibolous Terms in Aristotle, Arabic Philosophy and Maimonides," 
Harvard Theological Review, Vol. 31, 1938, pp. 151-73). Aristotle in his 
"catsgoriss" gives man and' his portrait as an example of an equivocal 
imushtmrmh) term for they both claim to being "animal". An Arabic 
example is the term " *ayn" which means both "eye" and "spring". 


Chapter Vll 


It concerns the general properties of the attributes 
— what is common to them all and that in which they 
differ. These are four l00m properties. 

The first property. The seven attributes which we 
have proved are not the essence [i.e. not God's essence], 
but rather they are distinct and superadded to the 
essence (dhat). 1 We maintain that the Maker of the 
world is knowing according to knowledge, living 
according to life, powerful according to power, and so 
is the case in the rest of the attributes. The Mu'tazilab 
and the philosophers deny that. They maintain that 
the eternal is "one essence" and that it is not per- 
missible to affirm multiple eternal essences 10u They 
also maintain that the proof points to this being 
knowing, powerful and living, not to knowledge, life 
and power. Let us single out "knowledge*' so that we 
do not have to repeat all of the attributes. They 
[Mu'tazilah and the philosophers] claim that know- 
ingness [being 'alim] is a mode (halah) 1 of the essence 
(dhat)— not an attribute ; but the Mu'tazilah oppose 
[the philosophers] in two attributes when they say 
that He is willing according to a will superadded 
to the essence and is speaking according to speech 
which is superadded to the essence except that He 
creates the will not in a locus and creates the speech 
in an inanimate body 3 and "thus" He becomes speak- 
ing according to it. The philosophers pushed ahead 


AlGhatali on Divine Predicates 

their analogy to "the will" and as for speech, they 
say He is speaking in the sense that He creates in 
the essence of the Prophet the hearing of arranged 
sounds either in [the Prophet's] sleep or in his state 
of consciousness, and that these sounds would have 
absolutely no existence outside the essence. Rather, 
[these sounds exist] in the hearing of the Prophet 
just as a sleeping person would see non-existent in- 
dividuals though their forms occur in his mind 
(dimagh). He also hears non-existent sounds which a 
person present [at the side] of a steeping person does 
not hear while the person who is sleeping may hear 
[these sounds] and become terrified and disturbed and 
[hence] wake frightened and alarmed. 4 They [also] 
claim that if a prophet ranks high in prophecy, the 
purity of his soul leads to his seeing, in his state of 
consciousness, marvellous forms and he hears from 
them harmonious, sounds which he learns by heart 
while those around him hear and see nothing. 3 By 
[these marvellous forms] they mean the vision of the 
angels and the hearing of the Qur'an from them. But 
[a prophet] who does not rank high in prophecy sees 
these [forms] only in [his] sleep. This is the gist of 
the doctrines of the misguided. 

Our purpose is to affirm the attributes and the 
decisive proof is that, that which helps [in affirming] 
that God is knowing, helps in [affirming] that He had 
knowledge because what is understood from our say- 
ing "knowing" and "has knowledge" is one and the 
same thing, since the intellect 102 * intellects an essence 
and intellects it in a mode and as an attribute. There- 
fore, it [i.e. » the intellect] would be intellecting the 
attribute and what is described. The attribute,"know- 


The Second Section of this Cardinal Point 

ledge," for example, is expressed in two statements. 
One of them is a long statement like our saying : In 
this essence (dhdt) knowledge subsists. 

The other is concise and short. It is being made 
concise by morphology and derivation, viz. "The 
essence is knowing," just as man sees a person and 
sees a shoe, and he [man] sees his [the person's] foot 
entering into the shoe. He [man] may express what 
he has seen in a long statement by saying: "The foot 
of this man is entering into his shoe/' or, he may 
express it by [simply] saying: "He is shod." The only 
meaning of his "being shod" is that he has a shoe. 
The Mu'tazilah conception that the subsistence of 
knowledge in the essence necessitates a mode called 
"knowingness" for the essence is pure infatuation be- 
cause knowledge is that mode. The only meaning of 
his being knowing is that the essence (dhdt) is [in] an 
attribute and in a mode. That attribute and that mode 
are the knowledge only. But those who derive mean 
ings from terms would no doubt err. And if the terms 
repeat themselves because of derivation, they would 
definitely err. The derivation of the term "knowing" 
('alim) from the term "knowledge" ('Urn) produced 
this mistake and, therefore, [people] should not be 
misled by it. Because of this all that which has been 
said and elaborated upon concerning cause and effect 
is refuted* The refutation of this is a priori to those 
whose hearing never experienced the repetition of 
those terms. As for those in whose understanding such 
terms stick, it would be possible only to erase them 
from their [heads] by a long tract which this brief 
treatise cannot bear. 

The concluiion We ask the philosophers and the 
Mu'fazilah : Is what is understood from our saying 


Al-Ghaz&li on Divine Predicate* 

"knowing" ('Slim) different from our saying existent 
(mawjud), or is there what points to the existence in it 
of something additional? 

If they say, "no," then he who says a "knowing 
existent" is like his saying an "existent existent" and 
this is obviously absurd. But if there is something 
additional to its concept, is it [i.e. this addition] 
specific to the essence of the existent or not? If they 
say, "it is not," this would be absurd, because then it 
would have nothing to do with its being an attribute 
to the existent [i.e. God]. But if it is specified by its 
essence Lie. the essence of the existent], then what 
we mean by "knowledge" is [exactly] that which is 
the existent specified addition which is superadded to 
the existent and from 103 * which it would be better to 
derive for the existent, because of it, the term "know- 
ing" ('Slim). This may help them 104 * understand the 
meaning. The dispute, therefore, becomes limited to 
the term and if I want to challenge the philosophers, 
I would say : Is what is understood from our spying 
"powerful" the same as our saying "knowing" ('Slim), 
or is it som thing else? If it is exactly the same, then 
it is as if we have said a "powerful powerful," which 
is sheer repetition. And if it is something else, then 
that is what we mean and you have, thereby, affirmed 
two meanings One of them is expressed by the term 
"power" (quirah) and the other by the term "know- 
ledge" {'um), thus bringing the whole dispute once 
more to the term. 

// it is said : 

Is what is understood from your saying, command, 
prohibition (amrwanahy) and information (khabar) 105 * 
the same as what is understood from your saying, 
commanding (amir), prohibitor (ndhin), and informing 


The Second Section of this Cardinal Point 

(mukhbir) or is it something else ? 

If these terms are the same, this would be sheer 
repetition and if they are something else, then He 
should have a speech which is command {amr) and 
another which is prohibition (nahy) and a third which 
is information and, furthermore, He should address 
every prophet lM * with a different speech. Likewise, is 
what is understood from your saying, "He knows the 
accidents," the same as what is understood from your 
saying, "He knows the substances (jawdhir)" or is it 
something else ? If it is the same, then the one who 
knows the substances (jawdhir) should know the ac- 
cidents through 107 * the same knowledge (Him) so that 
one knowledge may be related to multiple and infinite 
objects. But if it is something else, God, Most High, 
must have varioLS infinite knowledges (*u/«m). Like- 
wise the [divine] speech, power, and will, and all those 
attributes whose relations to their objects are infinite 
should, therefore, be infinite in numbers, which is 
absurd. If it is permissible that there be one attribute 
which h at the same time the "command," the "pro- 
hibition," and the "information," and would serve to 
replace these different terms, then it should be permis- 
sible that there be one attribute which replaces know- 
ledge, power, life, and the rest of the attributes Con- 
sequently, if this is permissible, it should be permis- 
sible that the essence (dhdt) be sufficient by itself and 
that therein lies the concept (ma'nd) of power, know- 
ledge, and the rest of the attributes without anything 
being superadded, thus making the doctrine of the 
Mu'tazilah and the philosophers a necessary corollary. 

The answer is to say : 

This question stirs the greatest problem of attri- 
butes and it would be inappropriate to solve it in 


AUGkai&li on Divine Predicates 

a brief treatise. But since the pen has already slipped 
to its mention, we may as well hint at the starting 
point on the way to its solution, though it is a ques- 
tion which most of the faithful' 08 " have avoided' * 
by reverting to the adherence to the Book (Qur'an) 
and the consensus. They [i.e. the faithful] say that 
the Divine Law {Shar*) cites these attributes because 
it points to knowledge ('ilm) and from this the one- 
ness [of knowledge] is undoubtedly understood. No 
addiiion to the oneness has been cited, and, therefore, 
we do not believe it. But this does not seem to be 
satisfactory because {Shar ') has cited command, pro- 
hibition, information (amr-nahy-khabar), the Torah, 
the Bible, and the Qur'an ; what, then, prevents us 
from saying that command is not prohibition, that 
the Qur'an is something other than the Torah, when 
it is cited that He, Most High, knows what is secret 
and what is open, the exoteric and the esoteric, the 
moistened and the dry, and so forth to all that which 
is included in the Qur'an The answer to this question 
may be what we should point to the beginning of its 


And that is that any intelligent group is forced to 
acknowledge 1 ' * that the proof points to a matter dis- 
tinct from the existence of the essence of the Maker, 
and this is what is expressed by His being knowing 
('*<**), powerful, and the others. This bears three 
probabilities : two extremes and a middle. Modera- 
tion, however, is the closest to accuracy. 

As for the two extremes, one of them shows care- 
lessness and that is the confining of oneself to think 
of one essence which expresses all these concepts 
and replaces them. This is the position of the philo- 
sophers! The second extreme is on the excessive side 


Tie Second Section of this Ctrdine* Point 

and that is the affirmation of an attribute with in- 
finite individuals such as knowledges, powers, and 
speech parallel to the number of the things related to 
these attributes. This is an excessiveness into which 
only the Mu'tazilah and some of the Karramites 
would delve. 8 The third opinion, which is our inten- 
tion, and which is the middle course, is to say that 
different things have levels in proximity and re- 
moteness in their differences. Two things may differ by 
their essences like the difference between motion and 
rest, nu between power and knowledge, between the 
substance and the accident. And whereas two things 
may come under the same definition and reality, they 
do not differ because of their essences but rather their 
difference is due to the difference of their relation. 4 ' 2 " 
Indeed, the difference between power and knowledge 
is\not like the difference between the knowledge of 
one blackness and the knowledge of another kind of 
blackness or whiteness, and this would be so if you 
define knowledge by a definition in which the know- 
ledge of all the objects of knowledge would .enter. 

We, therefore, say that the golden mean in belief 
is to say that any difference is due to the difference 
of the essences in themselves. One of them would not 
be sufficient to replace the other? which are different 9 
Therefore, it is necessary"* that knowledge be some- 
thing other than power, and so are life and thi rest 
of tfce seven attributes. It is also necessary that the 
attributes be something other than the-essence inas- 
much as the difference between the described essence 
and the attribute is stronger than the difference 
between any two attributes. 

As for the knowledge of a thing, it does not differ 
from the knowledge of another except in the aspect 


Al-Gkaidll on Divine Predicates 

of its relation to the object of relationship It is not 
remote that the eternal attribute be distinguished by 
this characteristic, viz. the difference in things relat- 
ed should not necessitate a difference 114 * and multi- 
plicity in it. 
If it is said : 

Nothing in what you have said eliminates the 
problem because if you admit a certain difference be- 
cause 115 ' of the difference of the object of relation, the 
problem would surely remain. Why do you investi- 
gate the cause of the difference after seeing that the 
difference does exist ? 
/ say ; 

The purpose of a supporter of a specific doctrine 
is to shqw, beyond question, the preponderance of his 
belief over that of the others and this we have defi- 
nitely demonstrated since we have shown that there is 
only one course to follow among these three, or the 
invention of a fourth unintelligible course [would 
be further required] This only course, when contrasted 
with its two opposite 116 ' extremes, its preponderance 
would be known definitely. If it were inevitable that 
it should be believed— and only one of these three 
could be believed ; ours being the closest of all three 
[to accuracy]— therefore, it should be believed. What 
remains to be considered are the implications that are 
required of this [third course] though the implications 
of the other two are even greater. Explaining 1 * 7 ' away 
the implications of [our course] is possible but eliminat- 
ing the difi&culty entirely, 118 * while the subject under 
investigation is the eternal attributes of God which 
are beyond man's understanding, is an inscrutable 
task within the scope of such a book. This, therefore, 
is all we can say here. As for the Mu'tazilah, we shall 


The Second Section of this Cardinal Point 
take them to task over their differentiation between 
power and will. 

We say : , . . . . 

H it is possible that He be powerful without power, 
it would bVpermissible that He be willing withoutw.ll. 
and there would be no difference between the two. 
If tt is said: 

He is powerful in virtue of Himself [6. nafsthi]" 
and. therefore. He is powerful over all the objects of 
power and. likewise, if He is willing in virtue of Him- 
self He would be willing the totality of the objects of 
will- this would be absurd because the contradictory 
objects of will could be willed only in terms of either- 
or. not collectively, while power can possibly be 
related to two contradictory things. 

The answer is: . 

You should say that He is willing m virtue of 
Himself, then becomes characterised by some of the 
contingent objects of will just as you have said that He 
is powerful unto Himself and that His power is related 
to only some of the contingent things because accord- 
ing to you. theentire acts of animals and the generated 
things lie outside both His power and His will If this 
,s possible vis-a vis power, it should be possible for 

"'Vs" for the philosophers, thev have fallen into con- 
tradictions concerning the attribute of speech. Their 
oosition is invalid from two aspects : 
P First, they say that God is speaking though" 
thev neither affirm the inner speech [kalamu-nnafsj 
nor do they affirm the existence of sounds externally 
They, however, affirm the hearing of sounds which 
they hold are created in the ears of the Prophet, 
peace be upon him, without any external sound. If it is 

Al-Ghat&H on Divine Predicates 

permissible that He be described as speaking through 
that which occurs in the mind (dimagh) of someone 
else [i e in the Prophet's mind]/ 20 * it should be per- 
missible that He be described as sound-producing and 
moving through the existence of sound and motion in 
someone else, which is absurd. 

Secondly, all that which they have mentioned is a 
rejection of the Divine Law in toto because what is 
perceived by a sleeping person is a phantasm, not real. 
If the Prophet's knowledge of the Divine Speech is at- 
tributed to imagination which is like confused dreams, 
the Prophet would not trust it and it would not be 
knowledge. On the whole, those [i.e. the philosophers] 
do not believe in religion and Islam. They merely em- 
bellish themselves with certain statements which they 
pronounce as a cautionary measure against the sword 
[i e. for fear of punishment]. The issue we have with 
them concerns the basis of the acts, the contingency 
of the world and the divine power {qudrah). We should 
not go with them into other [distracting] details. 

Ij it is saii : 

Do you maintain that the divine attributes are 
something other than God ? 

We say : 

This is wrong. When we say "Allah," we point to 
the essence (dhai) together with the attributes, not to 
the essence alone because the term "AUdft" could not 
be predicated of an essence that is judged to be free 
from the divine attributes" 1 ' whereas it could not be 
said 1 "* that jurisprudence is something other than 
the jurist ; Zayd's hand is something other than Zayd, 
and the carpenter's hand is something other than the 
carpenter because part of what is included in the 
name would not be something other than what comes 


r he Second Section of this Cardinal Point 

under the name. Therefore, the hand of Sayd is 

neither Zayd himself nor is it something other than 

Zayd. Rather, either way you posit it would be 

absurd. Likewise, any part is neither something other 

than the whole nor is it the whole itself. If it is said 

that jurisprudence is something other than man, this 

is permissible, but it is not permissible to say that it 

is something other than the jurist because man is no 

indication of the attribute of jurisprudence. It is small 

wonder, therefore, that it be certainly permis^ble to 

sav that the attribute is something other than the 

essence in which the attribute subsists^ Likewise it 

can be said that the accident that subsists in the 

essence "jawhar", is something other than thV'jawhar, 

meaning that the concept [of the term accident] is 

something other than the concept [of the term essence]. 

This is possible on two conditions : 

(!) That the Divine Law (Shaf) does not prevent 
its use and this is limited to God alone. 

(2) It should not be understood from the term 
'•other" that whose existence is possible to the exclu- 
sion of that which is other than it in relation to it. 
If this is understood, it would not be possible to say 
that Zayd's blackness is something other than Zayd 
because [blackness] does not exist without Zayd. 
Therefore, the deanition of the concepts and the 
terms is being made clear and there is no sense in 
dwelling on what is self-evident. 


J££7Z* »* « M. «. Abu *.i ».-J.«i. The, ** oot ...y 


Al-Ghaidli on Divins Predicate 

absolutely the exigence of attributes bat rather denied their subsistence 
in the essence as something superadded to it. affirming, however, their 
existence as something identical with the essence, the kiod of existence 
hich Abu Hlshim ascribes to what he calls modes (ahwil). {Kitdb Jabaqit 
^l-Mutatilah, p 96. ed Susanna Diwald-Wilzer, Beirut, 1761 : Badtdi, 
pp 18.93,169.95. 111.144. 

3. See note I, Chapter VI. 

4 Ibn Sini reduces speech to a kind of dream {Al-Najdt,, pp. 157-93). 

5 ' Ibn Sioi does not differ very much on this point .See ibid., pp. 157-93, 

6. See Introduction and our Chapter II. 

7 philosophers and some of Mu'tazilah are agreed that God's essence 
expresse all these attributes. See Introduction and our Chapter II. 

g. See our note on Karramites. 

9. It is Ghazali's contention that each attribute has a different essence 
and hence kind of relation and, therefore, all attributes must be different 
from each other. This he is directing against al-Ka'bi who reduced every- 
thing to the attribute of knowledge. 

10. Wolfsoo discusses the terms bi-nafsihi and li-dhatihi in his article 
already mentioned in the Honunaje A At Mas Vallicrosa, Vol. llj pp. 545-71. 


Chapter VIU 


We maintain that all the attributes subsist in His 
essence and none of them could possibly subsist with- 
out His essence, whether [this attribute] is in a locus 

° f n0t * ... Ml 

As for the Mu'tazilah, they maintain that will 
does not subsist in His essence and, therefore, it is a 
contingent thing {hadtthah)* that He is not a locus 
for contingent things, that [will] does not subsist in 
another locus because that would lead to His being a 
wilier through that locus, and that [will] exists in no 
locus. They also claim that the divine speech does 
not subsist in God's essence because it is a contingent 
thing and, therefore, it subsists in a body (jism) which 
is inanimate so that He would not be speaking through 
it but, rather, the speaker is God, the Most Exalted. 

As for the proof that the attributes shourd subsist 
in the essence, it would not be needed by those who 
have understood what we have presented above. Since 
the proof indicates the existence of the Maker, it 
necessarily indicates that the Maker is of such and 
such attributes and we mean nothing by-His being of 
such and such attribute other than that He possesses 
that attribute. There is, surely, no difference between 
His having an attribute and the subsistence of that 
attribute in His essence. 

We have already made clear the meaning of our 
statement that His being knowing is the same as say- 


Al-Ghaxili on Divine Predicates 

ing that there is knowledge in His essence, which is 
similar 123 * to the meaning of our statement that His 
being willing is the same as saying that will subsists 
in His essence. And, likewise, the meaning of our say- 
ing that will does not subsist in His essence is analogi- 
cal to our statement that He does not will. Calling 
the divine essence "willing" through a will that does 
not subsist in it is like calling Him "moving" through 
a movement that does not subsist in Him. If will does 
not subsist in His essence, it will make no difference 
whether it is existent or non-existent. Therefore, the 
statement of those who say that He is "willing" is a 
meaningless, erroneous term ; and likewise, so is the 
term "speaking". He is speaking because of His being 
a locus for speech and, therefore, there is no difference 
between our statements : He is speaking, and speech 
subsists in Him. Likewise, there is no difference be- 
tween our statements : He is not speaking, and speech 
does not subsist in His essence. This is exactly ana- 
logical to His being "sound-producing" and "mov- 
ing". If our statement, speech does not subsist in His 
essence, is predicated of God, thus our statement, He 
is not speaking, would be rightly predicated of Him, 
because both 124 * statements are an expiession ol one 
meaning. The unheard-of thing is their doctrine that 
will exists in no locus. If it is possible that one of the 
attributes would exist not in a locus, then it should 
be possible that knowledge, power, blackness, motion, 
and even speech, can exist in no locus. For that 
matter, why do they single out sound as being created 
in a locus ?, let it be created in no locus too. it sound 
cannot be conceived ot except in a locus because it is 
an accident and property, likewise is will (irduah). 
Conversely, you say that if He creates speech in no 


The Second Property of the Attributes 

locus and will in a locus, the thesis and antithesis will 
be the same. 

But since the creation of the first creature requires 
will and since the locus is created, they could not 
judge the locus of the will to be existent before the 
will because there is no locus before will except God's 
essence {dhat). But they do not make Him a locus for 
contingent things though those who think of Him as 
a locus for contingent things are closer [to being 
sensible] than they are. Therefore, the absurdity of 
the existence of a will in no locus, the impossibility 
of His being willing through a will that does not 
subsist in Him, and the absurdity of the occurrence 
of a contingent will through no will, 125 * can be per- 
ceived axiomatically by the intellect or by its sober 
investigation. These are three evident absurdities. As 
for the absurdity of His being a locus for contingent 
things, this could not be perceived except through 
detailed logical demonstration as we shall make clear. 


l. See notes on Will and Sp$sek. 


Chapter IX 

All the attributes are eternal, because if they were 
originated, the Eternal a parte ante would be a locus 
for contingent things which is absurd, or He would be 
qualified by an attribute which does not subsist in 
Him, and this is even more absurd as has already been 
mentioned. ' Nobody ever claimed the origination of 
[the attributes] "Life" and "Power**. They do believe, 
however, in "the origination*' of the knowledge of con- 
tingent things, divine will, and divine speech. 2 We 
point to the absurdity of His being a locus for contin- 
gent things from three perspectives. 

First, every contingent thing can possibly exist. The 
sempiternal a parte ante is a Necessary Being ( Wdjibu- 
l-Wujud) and if His attributes were [objects] of possi- 
bility, this would be contrary to His necessary existence 
because possibility and necessity are two contradictory 
matters. 5 It is impossible for a Necessary Essence 
(W&itbudhDhat) to possess possible attiibutes. This 
is self-evident. 

The second aspect is the strongest of all three If 
it could be supposed that contingent things reside in 
[God's] essence, it would not be without : either the 
imagination would ascend to a contingent thing, the 
existence of a contingent thing before it being impos- 
sible, or it would not ascend. Nay, before every 126 * 
contingent thing it should be possible that there be a 
contingent thing. If the imagination does not ascend 
to this, then His qualification by contingent things 


The Third Property 

eternally would necessarily be possible. This would 
eventually lead to a chain of contingent things that 
have no beginning and the proof of its impossibility 
has been furnished. No sane person has maintained 
this part of the argument. But if the imagination leads 
to a contingent thing, 127 * the origination of a contin- 
gent thing before it being impossible, then that impos- 
sibility of the originated thing, being receptive to the 
contingent thing in itself, would not be without its 
being an impossibility either because of the contingent 
thing per se or because of something distinct from it 
[i.e. because of something distinct from the hadith]. 
It is absurd, however, that it be because of something 
distinct from it, because every distinct thing that may 
be supposed [to exist] its non-existence could also be 
supposed, which leads to a chain of contingent things 
ad infinitum; and it is absurd. What remains, there- 
fore, is that the impossibility of this is in view of the 
thesis that the Necessary Being is qualified by an at- 
tribute that makes impossible the reception of contin- 
gent things because of His essence. If this is impos- 
sible ab eterno vis-a-vis His essence, then it would be 
absurd that the impossible be transformed into pos- 
sible, and this is on the same level as the impossibility 
of His being receptive to colour [i.e. mutation], eter- 
nally—for that [i.e. reception of colour] characterises 
mortals because "God," according to His essence, is not 
receptive to colours [i.e. mutation] by the consensus 
of all wise men. That impossibility cannot be trans- 
formed into possibility ; and likewise it is impossible 
that any contingent thing may reside in His essence. 

// it is said : 

This is negated by the origination of the world 



Al-Ghaxili on Divine Predicates 

because it was possible [mumkin] before its occurrence 
and the imagination cannot ascend to a time before 
which the occurrence of the world would be impossible 
although its occurrence within eternity is impos- 
sible. On the whole, we can say that" its occurrence is 
not impossible. 
We say : 

This objection is false. We declare only as impos- 
sible the affirmation of an essence (dhat) which 
shrinks from the acceptance of a contingent thing 
(hddith) because it is a necessary being, then changes 
to the possibility of accepting contingent things. The 
world does not have an essence (dhat) before coming 
into being which can be qualified as either receptive 
to origination, or not receptive, so that it may change 
to the acceptance of the possibility of origination so 
much so that this objection to our argument would be 

Yes : this is required of the Mu'tazilahs' position 
when they say that the world has a (dhat) whose 
non-existence li8 * (*adam) is eternal and which is recep- 
tive to origination which then occurs to it after it was 
not. According to our premise, this is not consequential. 
Our thesis is that the world is an act ; that an eternal 
act is an absurdity because the eternal cannot be 
an act. 

Thirdly, we say : 

If we assume the subsistence of a contingent thing 
in His essence, He— before that contingent thing- 
would either be described by the contrary of that 
contingent thing, or He would be described. as being 
separate irom ttiat contingent thing. 129 * if either that 
contrary or that separation were eternal, none of them 


The Third Property 

would be negated or made to cease to exist because 
eternal is incorruptible [i.e. is not an object of non- 
existence], and if either were a contingent thing, then 
there must be a contingent thing before it and before 
that contingent thing there must still be another ad 
infinitum which is absurd. This [absurdity] can be 
seen posited in a certain attribute such as the divine 
speech (kalam), for example. 

As for the Karramites, 4 they maintain that He is 
speaking in sempiternity, meaning that He is power- 
ful over the creation of speech in His essence,.and that 
whenever He originates something in other than His 
essence (dhat), He originates in His Essence His state- 
ment, "Be \"(Kun). He, therefore, must be silent before 
the creation of this statement and His silence, thus, 
would be eternal a parte ante (qadim). If Jahm, 5 how- 
ever, maintains that God creates in His essence know- 
ledge (Him), He must surely be, before its creation, 
unaware of it [i.e. unaware of this Him] and therefore, 
His unawareness would be eternal. 

We say : 

Eternal silence and eternal unawareness could not 
possibly be negated in view of the aforementioned 
reasoning, viz. the impossibility of the annihilation of 
the eternal. 

// it is said : 

Silence is not a "thing" (shay') f but rather is the 
privation of speech, and unawareness is the privation 
of knowledge, ignorance and its other opposites. When 
[divine] speech exists,, no "thing" (shay') is negated 
because there is no "thing" (shay') but the eternal 
essence which is eternal a pane post (bdq t yah), but 
another existent is addtd to it, which is speech and 


Al-Ghax&ll on Divine Predicates 

knowledge. But if it be said that a "thing" (shay*) 
has become non-existent [because of the occurrence 
of speech], this is unacceptable. This is on the same 
level i3 °* as that of the "existence of the world" for 
it negates the "eternal non-existence". But "non- 
existence" ('adam) is not a "thing" (shay)* so that it 
may be described as being "eternal" and, therefore, its 
negation supposed. 

Our answer to this claim is from two perspectives. 

Firstly, those who claim that silence is the priva- 
tion of speech and is not an attribute (sifah), that 
unawareness {ghuflah) is the privation of knowledge 
and is not an attribute, is like their claim that the 
colour "white" is the non-existence of the colour 
"black" and the rest of the colours, and is not a 
colour ; that "rest" is the non-existence of "motion" 
and is not an accident. All this is absurd, for the 
very proof that points to the absurdity of the first 
[part of their claim] points to the absurdity of the 

second [part]. 

Our adversaries on this question acknowledge that 
"rest" is an attribute whichismorethanjust theabsence 
of "motion". Indeed, whoever claims that "rest" is the 
non-existence of "motion" [alone] will not be able to 
affirm the origination of the world. Therefore, if l,u 
the appearance of motion after rest points to the ori- 
gination of the moving thing ; so wouid the appear- 
ance of speech (katam) after silence p>int to the ori- 
gination of the speaker," 2 * and no difference [between 
the two cases can be defended], because the very 
method employed in discerning that rest is a concept 
contrary to motion can be employed to show that 
silence is a concept contrary to speech, and that un- 


The Third Property 

awareness is a concept contrary to knowledge. If l33a we 
perceive a difference between the two modes of an 
essence : when resting (sdkinah) and when in motion 
(mutator rikah) ; then the essence is perceivable in 
either mode and the difference between the two modes 
is perceivable. Now this difference is due to nothing 
but the disappearance of something and the occur- 
rence of another because a "thing" (al-shay*) does 
not differ from itself which indicates that any re- 
cipient of a "thing" would not be free from that thing 
or of its opposite. 134 * This is applicable to speech 
(kalam) and knowledge. But it is not a corollary that 
there be a difference between the existence and non- 
existence of the world 133 * because in the case above, 
the existence of two essences was not a concomitant, 
while [in the case of the world] we do not perceive 
in both modes [i.e. in the modes of 'adam and wujud] 
one single essence to which existence occurs. Nay, the 
world has no essence before occurrence, while the 
eternal a parte ante has an essence before the occur- 
rence of speech which is known in a mode different 
from the mode in which it [i.e. the essence] is known 
after the occurrence of the speech. The former [i.e. 
before occurrence of speech] is termed "silence" and 
the latter is termed "speaking" These are t*o dif- 
ferent modes in which one-continuously-existing dhat 
is perceived. The essence has a particular mode of 
being, an attribute and a mode by which it is "silent," 
just as it has a mode and an attribute by which it is 
"speaking," and it also has a being by which it is 
"resting," "moving," and "black and white". This 
parallelism is inescapably consistent. 

Secondly, to finalise this argument [we propose 


AUGhatZUl on Divine Predicates 

that], if it is acknowledged that silence is not a [posi- 
tive] concept but rather is due to an essence that is 
separate from speech, then separateness from speech, 
surely, is a mode for the separate which would become 
non-existent 136 * once speech occurs. This mode of sep- 
arateness, be it called non-existence, existence, attn 
bute, or a particular mode of being {hay'ah) would be 
negated by speech. But the one negated is eternal and 
we have already mentioned that the eternal (qadim) 
cannot be negated, whether it is an essence (dhat), a 
mode (halah), or an attribute [sifah). The impossibi- 
lity [of its negation] is not because it is an essence 
alone j it is because it is eternal. To this one cannot 
object by the example of the non-existence of the 
world on the ground that it also became annihilated 
despite its eternity," 7 ' because the non-existence of 
the world is not an essence nor does it effect a mode 
for an essence, a mode which may be judged to alter- 
nate manners on the essence. The difference between 
the two is clear. 
If it is said: 

The accidents are numerous, and the opponents 
do not claim that the Creator, Praise be to Him, is a 
locus for the origination of any of them, such as co- 
lours, pain, pleasure, and others. The argument, how- 
ever concerns the seven attributes which you have 
mentioned. There is no dispute in the attributes of 
"life" and "power," but rather the dispute is on the 
three attributes of speech,™* will, and knowledge j in- 
cluded in the concept of knowledge are the attributes 
of "hearing" and "sight" amongst those who affirm 
them, that these three attributes should be originated. 
It is impossible that they subsist in other than Him 


The Third Property 

because then He would not be described by them and, 
therefore, they should subsist in His essence which 
necessitates His being a locus for contingent things. 

Concerning the knowing of contingentthings, Jahm 
maintains that it is a contingent attribute 7 because 
Allah Most Htgh. knows "now" that the world has 
existed before this [moment], that if in sempiternity, 
God knew that the world had existed, this would have 
been ignorance not knowledge, that if He did not 
know 139 * and "now" knows, then the contingency of 
knowledge would be shown by the fact that the world 
had existed before this [moment]. This applies to all 

As for will (iradoh), Jahm maintains, it should be 
originated because if l4 °* it were eternal, the object of 
will would be equally eternal with it, for whenever 
power and will are consummated, and all obstacles 
are done away with, the object of will should occur ; 
how he asks, could the object of will and power be 
delayed when there is no obstacle ? For this reason the 
Mu'tazilah claim the origination of will in no locus 
and the Karramites claim its [i.e. will's] origination in 
His essence. I he Karramites may express this by 
saying that He created an act of "creation" {ijau) in 
His essence at the time of the origination of every 
existent thing and that is due to will. 

As for speech (kaldtn), they ask, how could it be 
eternal when it contains information about things past? 
How they further ask, could He say in eternity, "We 
sent Noah to his people," when He did not create 
Noah yet? And how could He say in eternity, unto 
Moses, -Put off thy shoes, thou art in the holy vall «y» 
when He did not create Moses yet? And how could He 


Al-Ghaztlt on Divine Predicates 

command and prohibit without there being an object 
of command or of prohibition ? If this is absurd, then it 
became acknowledged by necessity that He commands 
and prohibits, such command and prohibition being 
impossible in eternity, 141 * then it would be decidedly 
known that He became Commander and Prohibitor 
after He was not. This is what is meant by His being 
a locus for contingent things. 
The answer. We say : 

When we solve this sophistry over these three at- 
tributes, there arises automatically an independent 
proof which negates His being a locus for contingent 
things. No one held such a view but for this sophistry 
which once examined would prove false, just as the 
theory which maintains that He is a locus for colours 
[i.e. mutation] and other [accidents] which no proof 
can justify His being described by them. 

We say, therefore : 

The Creator, Most High, knew in sempiternity of 
the existence of the world at the time o*f its existence. 
This knowledge is one attribute which requires decree- 
ing in sempiternity the knowledge that the world 
"would be" and at the time of its existence, that it 
"is" and after, the knowledge that it "was". These 
modes occur consecutively on the world, and the 
world is always unveiled to God through this attribute 
while He does not change. 142 " Rather, what changes 
are the modes of the world. To illustrate this, we say : 
Let us suppose that one of us has acquired a know- 
ledge before sunrise that Zayd would be coming at 
sunrise ; that this knowledge did not cease but remain- 
ed j that this one did not acquire any other knowledge 
at sunrise j what would the state of this person be at 


The Third Properly 

the rise of the sun ; would he be knowing that Zayd 
is coming or would he not be knowing ? It is absurd 
that he should not be knowing because the remaining 
of knowledge is judged by [Zayd's] coming and at 
sunrise. I4J " Now that he knows that the sun has risen, 
he would necessarily be knowing of the coming [of 
Zayd]. If 'this knowledge continues 144 " after sunrise, 
he should certainly be knowing that [Zayd] has come. 
Therefore, one knowledge provides the comprehen- 
sion that something is going to "be," it "is," and it 
"became". This example should help 145 * to under- 
stand God's eternal knowledge that necessitates the 
comprehension of contingent things in their different 
modes. Hearing and sight should be analogical to this 
because each 146 " one of them is an attribute by which 
the objects of sight and of hearing are made "seen" 
and "heard" clearly at the time of [their] existence 
without either attribute being originated or under- 
going change, Indeed, the contingent things are the 
objects of hearing and of sight. 147 * 

The decisive proof of this is that the difference 
between the modes of a single thing, 148 " in its categori- 
sation to that which has "become," ''shall be/' and 
"is," could not exceed the difference between the 
different essences. If it is, of course, acknowledged 
that knowledge 149 " does not become multiplied by the 
multiplicity of the essences, how, then, could it [i.e. 
knowledge] be multiplied by the multiplicity of the 
modes of one single essence ? And if one knowledge 
provides the comprehension of different and dissimilar 
essences, why should it be impossible that one know- 
ledge would provide comprehension of the modes of 
a single essence in relation to the past and the future ? 


Al-Ghaz&H on Divine Predicates 

There is no doubt that Jahm does not attribute 
limitation to God's objects of knowledge and that he 
does not affirm infinite knowledges; therefore, the 
consequent of this is that he should admit one know- 
ledge that is related to numerous and different objects 
of knowledge. How could it be improbable with respect 
to the modes of a single object of knowledge which 
He actualises? If a [individual] knowledge about 
every [individual] contingent thing occurs to Him, 
then that knowledge would not be free 130 ' either from 
being an object of knowledge or from not being an 
object of knowledge. If it would not be an object of 
knowledge, this would be absurd because it is a con- 
tingent thing and if it is possible that there exists a 
contingent thing which He doesn't know although it 
is in His essence and it is more proper for Him to be 
aware of it, then that it is possible that He would 
not know the contingent things which are different 
from His essence is even more proper. But if it is an 
object of knowledge either he would be lacking an- 
other knowledge which in turn requires further know- 
ledge, ad infinitum, 151 * which is absurd, or He would 
know the contingent thing and the "knowledge" of 
the contingent thing with that same knowledge thus 
making knowledge one essence having two objects of 
knowledge, 132 ' one of them being its essence [i.e. of 
knowledge] 153 ' and the other being the essence of the 
contingent thing. This would unquestionably require 
the possibility of a single knowledge related to two 
different objects of knowledge. How, then,- could it 
not be permissible that there be one knowledge which 
is related to the modes of a single object of knowledge 
and the preservation of the unity of that knowledge 


The Third Properly 

and its freedom from mutation ? This is inescapable. 
As for will (irddah), we have already mentioned 
that its occurrence without another will is absurd, that 
its [will's] occurrence through [another] will would 
lead to infinite regress. [We have also mentioned that] 
the relation of the eternal will to contingent things is 
not absurd and that it is impossible that will be relat- 
ed to the eternal (qadim), that the world is not eternal 
because will is related to its origination, not to its 
existence in eternity . 15 " All this has already been made 
clear. Al-Karr&ml, too, maintains that He creates in 
His essence an act of "creation" {ijad)* at the time 
of the origination of the world and thus the origina- 
tion of the world is fulfilled at that time. Al-Karr&mi 
would be asked : And what specifies the contingent act 
of creation {ijad hadith) in His essence by that [parti- 
cular] time ? Would not that time in turn require 
another specifier ? They [the Karramites] would be 
faced in their position of Ijad with the same conse- 
quences of the Mu'tazilahs' position concerning the 
contingent will. And if some of them [i.e. some of the 
Karramites] maintain that that [ijad] is His saying, 
"Be" (Kun), which is a sound, 135 ' this would be absurd 
from three perspectives : 

One of these perspectives is the impossibility of tne 
subsistence of sound 156 " in His essence. 

The second is that His statement : "Be" (Kun) is 
originated also. If it occurs without His saying to 1 ' 
"Be" "Be " then the world might as well occur with- 
out His saying to it, "Be" (Kun) If His statement : 
"Be" (Kun), in order for it to occur, requires another 
statement, this latter statement would, in turn require 
a third one, and the third a fourth ad infinitum. 


Al-Gkax&ll on Divine Predicates 

Furthermore, we should not debate with those whose 
intellects lead them to maintain that He creates in His 
essence His statement, "Be" [Kun), every time He 
creates something, thus bringing together thousands 
over thousands of sounds every moment. It is, of course, 
acknowledged that the letter (nun) and the letter (kaf) 
cannot be pronounced simultaneously. The letter (nun) 
should succeed the letter (kaf) because the combina- 
tion of the two letters together is impossible and if 
they are combined together, but not in order, they 
would not make sense or constitute a speech. Just as 
it is impossible to combine two different letters, it is 
also impossible to combine two similar letters. And as 
it is unintelligible [to pronounce] instantly a thousand 
(kafs), it is also unintelligbte [to pronounce] the letter 
(kaf) and the letter (nun) simultaneously. It is better 
for such people to ask God, Most High, to bless them 
with some intelligence than to dabble in intellectual 

Thirdly , is His command , "Be"(Kun) ,an address 158 ' 
to the world when it is in the state of non-existence or 
when it is in the state of existence (wujud) ? If it is an 
[address to the world] when it is in the state of non- 
existence, then the non-existent does not comprehend 
a command and, therefore, how could it comply 
by existing in response to His command, "Be" ? And 
if [the command to the world] is when it is in the state 
of existence, the existent cannot be asked to "be." 
Consider, now, what God can do to one who goes 
astray from His right path, how one's feeble intellect 
leads one not to understand the meaning of God's 
Word : "His command, when He desires a thing, is to 
say to it 'Be,' and it is." and that God's Word is only 


The Third Property 

a metaphor denoting the efficacy of [the Divine] Power 
and its perfectior . These are led by wrong understand- 
ing to these infamies. We seek God's protection from 
disgrace and scandal on the Great Day of Fright, the 
day when the consciences are unveiled, the hearts 
afflicted, and the veil of God shall be removed from 
the viciousness of the ignorant, when the ignorant 
who hold wrong views about the belief in God and 
His attribute will be told, "Thou wast heedless of this ; 
therefore, We have now removed from thee thy cover- 
ing, and so thy sight today is piercing." 

As for speech (kalam), it is eternal and what they 
consider as improbable in His saying [to Moses], "Put 
off thy shoes," and His Word, "We sent Noah to his 
people," is an improbability which leans on their 
judging speech as sound, which is impossible to God, 
but [speech] is not impossible if the [theory] of "inner 
speech" is understood. 

We say : 

There subsists in God's essence information about 
sending Noah, the verbal expression of which, before 
sending him, is "We send," and after sending him, 
"We have sent." The term, naturally, differs accord- 
ing to circumstances but the concept that subsists in 
His essence remains unchanged. Its reality is that it 
is information related to an informer. That informa- 
tion is the sending of Noah at a particular time which 
does not change by the change in the modes as we 
have already mentioned when dealing w?th knowledge. 
Likewise, His Word, « Put off thy shoes," is a state- 
ment which indicates "command". The command is 
a requirement and a demand that subsists in the 
essence of the commander, and for it to subsists in the 


Al-Ghazdli on Divine Predicates 

commander's essence, the presence of the one com- 
manded is not a condition. But it is permissible that 
it subsists in his [i e. commander's] essence before the 
existence of the one commanded. If the one command- 
ed comes into existence, he would be 159m commanded 
by that same requirement without a further renewed 
requirement. For, many a person who has no son, as 
there would subsist in his essence a requirement for 
the quest of knowledge by his son, would estimate in 
his mind to say to his son : "Seek knowledge.'' This 
requirement would be effectuated in his mind on the 
assumption that the son would exist* If the son is to 
exist, and a mfnd is created for him as well as a 
knowledge of what is in the father's mind without the 
estimation of the composition of an audible statement, 
and if this requirement is to remain until the time 
when [the son] exists, then the son would know that 
he is commanded from his father's side to seek know- 
ledge without a need for the resumption of a require- 
ment to be renewed in the [father's] mind; rather 
that [initial] requirement remains. Yes, the custom is 
that a son would not have a knowledge [of what the 
father demands] except through a statement that in- 
dicates the inner demand [i.e. a statement that indi- 
cates what is in the father's mind]. The verbal state* 
ment, 160 ' "Seek knowledge," would indicate the de- 
mand that is in himself [i.e. in the father's mind], 
whether that demand occurs at that time when his 
son exists or it was subsisting in his [i e. in his father's] 
essence before his son's existence. This is, the way 
that the subsistence of command in God's essence 
should be understood; the statements indicating it 
[i.e. the command], being originated and the object of 


The Third Property 

indication [i.e. command itself] being eternal (qadim). 
The existence of the objection of indication does not 
require the existence of the one commanded; but 
rather it imagines his [i.e of one commanded] existence 
since it is possible that the one commanded can be 
estimated to exist. But if its existence is impossible, 
the existence of the requirement may not be imagined 
by one who knows the impossibility of its existence. 
Therefore, we do not hold that a requirement to effect 
an act from that the existence of which is impossible 
subsists in God's essence ; rather, the requirement 
should be directed toward that whose existence is 
possible; and this is not absurd. 

// it is said : 

Do you say that God, in eternity, is commanding 
and prohibiting? If you say He is commanding, how 
could He be commanding without there being one to 
command? And if you say, "No," then He becomes 
commanding after He was not. 

We say : 

Our friends [i.e. the Ash'arites] have differed on 
the answer to this question. Our choice [i.e. our view] 
is to say that this is an investigation, part of which 
is related to the meaning and the other part is relat- 
ed to the general application of the term linguistical- 
ly. As for the share of the meaning, it has been clari- 
fied, viz the eternal requirement is intelligible even 
thought it may precede the existence of the one com- 
manded as in the case of the son. Rather, it should 
be asked whether the term "commanding" 161 * should 
be applied to Him after the one commanded exists 
and understands or should it be applied to Him be- 
fore? This in any case, is a semantic question the 


Al-Ghaz&li on Divine Predicates 

likes of which should not occupy a theologian. The 
truth of the matter is that it is permissible to apply 
it to Him just as it has been permitted to call God, 
Most High, "powerful" before the existence of the 
objects of power. They do not consider as remote a 
powerful agent who does not have an existent object 
of power. Rather, they hold that the powerful requires 
an object of power that is an object of knowledge 
and which is not existent. Likewise, the commanding 
requires an object of command l6U which is an object 
of knowledge but not existent, while the non-exist- 
ent's existence is an object of knowledge before [its] 
existence. Nay, a command requires contents of com- 
mand, an object of command, and a commanding one, 
while the contents of command are non-existent. It 
should not be asked, how could He be commanding 163 * 
without contents of command, but it should be said 
that He has contents of command that are objects of 
knowledge and whose existence is not a condition j 
the condition is their non-existence. Nay, he who 
commands his son, by way of counselling, to do some- 
thing and then dies, after which the son executes the 
will, such son would be described as having complied 
with the command of his father though the command 
is missing here and the commanding one himself is 
non-existent. But in spite of this, we do apply the 
phrase, "compliance with command". If the one 
commanded is not complying with the command, and 
neither is the command existent nor the commanding 
one, and. furthermore, if that command being a com- 
mand 164 * did not require the existence of the contents 
of the command either, why should the existence of 
the one commanded" 3 ' be required? This much has 


The Third Property 

uncovered the share of both the term and the mean- 
ing which are our sole objects of investigation. This 
is the summation and analysis of what we have want- 
ed to say concerning His being a locus for contingent 


1. Some of the Mu'tazilah, as we have pointed out, hold that tome of 
the attributes are conti agent such as the attribute of will. 

2. See our note on Knowledge. 

3. In note 5, Chapter I, we have cited Ibn Sina's concept of al-Wijib. 

4. Ste note on KarramUes. 

5. Jabm ibn Safwan is the head of a school which carried his name (al- 
Jahmtyyah) He was a J abrite. See our note on Jabrites, and al-MiUl, p. 60. 

6. See Maqdttt, p. 181, and our note 22 in our Introduction. 

7. See note 5. Chapter IX. 

8. See our note on Karramihs. 


Chapter X 

The names of God, Most High, which are derived 
from these seven attributes, are sempiternally (azalin) 
and eternally predicated of Him. He is, in eternity, 
Living, Knowing, Powerful, Willing, 166 * Hearing, See- 
ing, and Speaking. As for those ones which are deriv- 
ed 167 * for Him from actions, such as : sustainer (rdziq), 
creator (khaliq), raising one (muHz), and lowering one 
(mudhil), opinions have differed 168 * on whether they 
are predicated of Him in sempiternity or not. But 
even these, if the cover is removed from them, the 
impossibility of the differences on them would be 

On the whole the names which God applies to 
Himself are from four categories :.* 

The first category points to nothing but His essence 
such as existent ; 169 ' and these are predicated sempi- 
ternally and eternally. 

In the second category are those which point to the 
essence together with a negative addition, such as [the 
name] "Eternal" a parte ante (al-qadi*n)r t which indi- 
cates an existence not preceded by non-existence, 410 * 
and 171 * the "Eternal" a parte post {baqi) which points 
to His existence [firstly] and the negation of non-exist- 
ence from Him secondly. Likewise are the "one" 112 * 
which points to the existence and the negation of a 
partner {shank) \ the "All-Sufficient" (ghani) which 


The Fourth Property 

points to the existence and negates need. Those, too, 
are predicated sempiternally and eternally because 
what is negated about Him is negated per se and such 
negation accompanies the essence perpetually. 

The third category encompasses those which point 
to existence plus an attribute which is superadded to 
it from among the real attributes [sifat al-ma'na] such 
as "Living," "Powerful," "Speaking," "Knowing," 
"Willing," "Hearing," and "Seeing," as well as what 
is related to these seven attributes such as "Con> 
manding," "Prohibiting," "Informing," 173 * and their 
like. All those, too, are sampiternally and eternally 
predicated of Him by those who believe in the eter- 
nity of all the attributes. 

The fourth category includes those which indicate 
existence in relation to an act of His acts such as the 
"Generous" (jawad), "Sustainer" (rdziq), "Creator" 
(khaliq) , "Raising One" (mWiz), "Lowering One" 
(mudhil) and their like. These are not agreed upon. 
Some maintain that they are predicated [of Him] 
sempiternally because if they were not to be predi- 
cated of Him, then His description by them would 
necessitate change. Others maintain that they are 
not to be predicated [of Him] because since there are 
no creatures in eternity, how could He be Creator ? 
What removes the cover from this is that the sword 
in its sheath is called sharp (s&Hm). -But at the 
moment of cutting it is called sharp by logical conclu- 
sion. But both cases have two different meanings. 
When the sword is in the sheath, it is sharp tn poten- 
tia, and when it cuts, it is sharp in actu. 

Likewise, the water in a glass and at the time of 
drinking is called thirst-quenching, yet these are two 


M-Gkaiill «• Divtnt Predicate 

Afferent applications. The meaning of calling the 
£S sWp'when it is in the sheath is that the 
characteristic by which the "cntting" is effected is 
Presence sword (» poUniia). The lack of imme- 
Sate cntting is not dne to a shortcom. ng ; > the 
lord's essence, or its sharpness or readiness to be 
Sbnt rather it is dne to a » f*"^ 
yond its essence. In the same sense that the ten* 
«,haro" is applied to the sword in its sheath, the 
t« m "Creator" (Utff) is applied to God in sempi- 
Sv Creation is not actualised because of some- 
Snewttat was not in the essence (dm )l rather 
StbZh conditional for the actualization of an ac 
!« «istent in sempiternity. Bnt in the sense that the 
erm S h predicated of the sword at the moment, 
„?™tZf r« e by logical conclusion] it is not predi- 
^ « Him^'n^mpiternity. This is the share of the 
^.i^ESS clear that those who say 

Wilted [of Him] in sempiternUy are right .f they 
iakeit in the second sense. And those who hold that 

Mrs tsrcsu ». «*- — 

^tribute of power, and five objections branched from 
^attribute" of speech. Amongst * — -ft 
perties of the , attnbut e* f our were^ ^ ^ 

*■■ J 100 

Tte Fourth Property 

claims. Every claim, however, was to be based on 
other claims which lead to its affirmation. 


1. See our Chapter II. 



AK: MS. No. 4129/1 in Library of Dil ve Tarih- 
Cogvafya Fakultesi, Ankara Universitesi. 

AS : MS. No. 2182 in the library of Ayasofiya. 
BA : 650 in Sulcymaniye, Be sir Aga. 


Note. — Numbers refer to the translation; the adopt- 
ed readings can be discerned by comparing the trans- 
lation with the variants and the 1962 Ankara edition 
of al-Iqtisdd FU-l'tiqSd. 

la AK-30, BA-140 oU^Jt J*l = AS-18 oU^JI M~T J„l 
2a AK 30, BA-140 ^-^ = AS-18 ^ 
3a AK-30, BA-140 J>UJ1 -uii.U*! J 

= AS-18 «j»ltJ» aIU*» j 
4a AK-30 ^toj- 4^ = AS-18 a*JI*iyu ^. 

« BA-140 ^1-ciyu^TjOJ 
Sa AK-30, BA-140 J*JI Sjj> *T,ju la* 

« AS-18 5jj>i <»^ ti* 
6a AK oSij ^uJjl AilJJ 

= AS-18. BA-140 **U JLSI3J jl 4JU1 
7a AK-30 J*Ui J*ij| ^ WJ = AS-19, ^.^ J-*>1 t^J I* 

= Ba-140 JiUJI J**JI l,j 
8a AK-3J, AS-19 jj-aJl ^ r ' ^ 
9a AK-31, BA-140 ol^ . AS-19 o^ 



10a AK-31 V 6i JU=J = AS-19 jJu*> VI JU*J' 

« BA-140 jJ^i ^ JUsJ 
11a AK-31, AS-19 lil = BA-140 il 
12a AK-31 «£^^ = AS-19 y>^j'- BA-140 £>> ^ 
13a AK-31, BA-140 ijJjfl - AS-19 Jj» 
14a AK-31, AS-19 J.W = BA-140 j^ 
15a AK-31 yjbjtji^ 
16a BA-140 ^ y» *H sv 
17a BA-141 .**** W^"j 
18a AS-19 ,JL»JI «- BA-141 ^Wl 
19a AS-19 *Jli ^ ^ kr** # 
20a BA-141 j 
21a AS-19 JUijl 
22a AS- 19 &JJI J >J» ^ jCJ U UU 
24a BA-141 & 

25a AK-33 J»Uy q\ ^Vj J UJ *>' 4> a & J*"*- 1 J 
26a AS-20 cW 1 Cr* 
27a AK-33 l>j-ju«JI 
28a AS-20 £Jtf jCs Vj 

29a AK-33 ^jx^JI - - - tfJ *l = AS-20 u-juJI t$j»i 
30a AS-20 ^Je-JI JlSCdUl 4jUJI JlCiVl wyl o' *tJ^tj 

31a AK-33 jk^tf 
32a AK-33 so**' Cx* J 1 * 3 
33a AS-20 aju-^JI [»*£&-•• j 
34a BA-142 >-jj 

35a AK-34, AS-21, BA-142 ^ jWI J' J» 
36a BA-142 j joi* cUl Jl UUJ crJj-tfJI 
37a AK-34 *yjJfij.V VI e^j 
38a AS-2W-BA-142 u«~ jj-^'j v«" •**> 
39a AS-21 , BA-142 *f-JI ,y ^ ^ 
40.1 BA-142 ^ol aIjUJI i^ijl oU 


Al-Ghaz&ll on Divine Predicates 

41a AK-35 «** jti ^x^tfS c J & 

= AS-21 h* cA^^ *** cfi v> *J U 

42a AK-35, AS-21 Ki jur ,/x. Sj^l <JU 

= BA-142 Ja* J*iii cM <-^o* W c r ■«• h^ oU 

43a AK-35 4>Vi J fJUb *iu< = AS^2l Jjbi J ^ *aut* 

44a AS-21 &* *4 >*™ ^AT j 

45a BA-143, W-> <•** 

46a AS-21 fJ ^ »-~JI Si 


48a AK-36 aT^LU ju>uu a** £^u 

49a AK-36 -51 jU* 

50a AS-22 ,JLM *. *^ 5*»j VI oft ^ 

51a AK-36 jU 

52a AK-36, AS-22, B A-143 6»jS» ^« o* >*& o» jj« 

53a AS-22 J ^>JljU $* J^tf c^l ifJli &*- & S J'* SJU 
^iJI .I^L ISU SajjJl ^ jy gUJI a-U .u* oJI 

54a AS-22 eM*» *# J>^ ** - BA-143 ji/«l <* j>^* 

55a AK-36 cJ> iSt 6l^VI ^ jj*^ » AS-22 

56a AK-37 ^j~i M U* 

57a AS-22 Sj^j jjoi^ ^ Lj 

58a AS-22 <JUdlj j*dlj* ,* 


60a AK-37 a^i* j*» •_** **U 6* -> 

61a AK-37 £h = BA-144 ^Ui 

62a AS-22 a^Ui* ^ Uu* # Vjt Uj»*. «ji 

63a AK-38, BA-144 L»jjUr J*« j &&J\ J&u J> 

= AS-23--- JWj---J^ul 
64a AK-38 aU ^e* d& 

= AS-23 ^UJl Jjft j-4ju aH' «^ 

.- BA-144 aU ^l d& aJU> Lj 
65a AK-38 oi>Ji Ij* l^ u**l U 
66a AS-23 ^ **Ij Jw £1 oil .uji ,JtJl 6' 


67a AS-23 i^*i ^ J ~V a"*"* ^ 1 * 4 ** ^ ^ U 

68a AS-24 J^l J* v> JI u" ^ 

69a AK-40, AS-24, BA-145 ** ^,M*> ** 

70a AS-24 Ji~ j* -> 
71a AK-40 *l~ jl = BA-145 U~ jl 
T2a AS-24 ok*> fa J 
73a AK-41 ^JWl j ^ «W» ^ 0* .-** 
74a AK-41 a^aJI 

75a AS-25 f l~^l ♦*> <^ ■> *~ Ju ^-> ^ J * J 
76a AK-42, AS-25 )M\ 
77a AS-25 j-UJI 
78a AS-25 J^" 

79a AK-42, AS-25, BA-146 JUT aJo~ ^ Sh^M tjs- 


80a AK-43, AS-26 jaJl C)U 

81a AK-43 J 31 ^ 1 A >^ v-'j #* 

82a AS-26 ^^' r^-^ 
83a AK-43 >Ji ,^=Jt jr 

84a AK-43, AS-26, BA-147 cJ^lj fJ L0l 6* ^J u ^ 

85a AK-43 f ^T a^U Jf\j = AS-26 a,^JU 

86a AS-26 Jtfrt ** jUy ^U UjI ^»j 

= AK-43 iiWt -^ r 1 * 
87a BA-147, AS-26 £>^» ^' ^j 
88a AK-44, AS-27, BA-147 j>*Ui s^ljl Jl LI 
89a AK-44, AS-27, BA-147 ^ j^' >W -»» 
90a AK-44 f j*t jl Jjj^j ^ 
91a AK-44 ^ N ax^**j 
92a AK-44 ^bt *^*^ *=^ J 
93a AK-44, AS-27 ^» j^> ^ y« ■ 
94a AK-44, BA-148 >J l J* o^iu »* 

= AS-27 j»»J*j-,..-f <ili 

95a AK-44 JUJI a- ^ A u J* ,jM i 


Al-G ha z&ll on Divine Predicates 

96a AS-27 JuT^y *M 

97a AK-47 djA^)\ .>y u 

98a AK-47 6MI 

99a AK-47 j>~ J* 
100a AS-29 f £*i «b* w *j 
101a AK-48 i*JJ^^ Atjjii olji 
102a AK-48 JfcuJ^i 6U 
103a AK-49 *^ 
104a AK-49 ^ju l^ j& 
105a AK-49 ^\ r fly 
106a AK-49 ^J Jf 
107a AS-29 y^l iiJij^ ^^jl 
108a AK-49, AS-29 iu*Jl = BA-150 <>*^i 
109a AK-49, AS-29, BA-150 X 
1 10a AK-50 lj^ 61 Ji >^* 
Ilia AK-50 il^*Jtj iT^Jl ui^Li 
112a AK-50 <>*Ji j& *t~ <y 
113a AK-50, BA = 151 ,JUH 6^. jl .^^J 
114a AK-50 ^U- ^-.y V jl > j 

a AS- 30 #U fc-**.>i V LJtfj 

1 1 5a AS-30 <J:teii c~J U J^ti 
1 1 6a AK-50 *J ^Wl «*> 
117a AK-50 d& VI jjaj j « BA-151 JKCtUi jj*j j 
118a AS-30 Wo *jrf ^ Jics VI ^Jtf j 
1 19a AK-51 AS-30 dy& V rf JV 
120a AS-30 *J>#>- ,J£* j^ tLa J 
=» BA-151 i»>*j-* *J^ Uo j 
121a AS-30, BA- 51 *-fJVi oli^JI 
122a AK-51 *^l> *2*u JUj VLT^ BA-151 4^1 ^ aUJI 
123a AK-52 ju^ uy r ^ j 
124a AK-52, BA-151 o* o&jU t^V ^ 
125a AS- 1 oljl :A* *oU ajtjl cjju 
126a AK-52, AS-31 JW *s»*U Jf J, 



127a AK-52, AS-31 JU 
128a AS-32 **»* ^ jo'i 

129a AK-53 xuiiJOi ji xJ» J3i j o^UJI *J0i ^ iJSii Vb jl. 

iifc: Vi *J3i y 
= AS-32 J£yyi j jujJ» «-&^ j MS is defective here 
130a AS-32 r >bJl ^j JJ* j * 
131a AS-32 *T j*M j*J* ^' **■>- 
132a AK-54 f *fli o-u 
133a AK-54 uT^«uij»j 

134a AS-32 13» j *** 6 fi -^ ** 
1 35a AK-54, AS-32, BA-153 (JbJl 

136a AK-54 r !*ai oU> r ^ 

137a AK-54 ^^' -fc* .J*** * u - AS-32 f-^l ^ i/-- 

138a AK-54, BA-153 r *&» 

139a AS-32 UU ^C ^ liij 

140a AS-32 ^ J& "^" ^£' 

141a AK-55 r J*J' 

142a AK-55 >*J1 Ui! j j^a ^ ^*j 

143a AK-55 £j^' ju* r jjau pLJi *u> >* 

144a AK-55 tilUi »Ual x* Jj >^j 

145a AK-55 M <M ^ c^ o* v/r* '-^ 

146a AK-55, AS 33, BA-153 L^ ^Ij JTuU 

147a AS-33, BA-153 \*jJIj *>—J» **H' 

148a AK-55, AS 33, BA-153 J»lj V cV Orf 

149a AK-55, AS-33 Ju£ o»ji)l ^xcj >*** V ^Wl 

150a AK-55, AS-33, BA-153 U jU*V ^Wl ^Ji o& 

151a AK-55 i«J *ilH V >• r>* Jl ^ 

152a AK-55 dk>u- *Jjl 

153a AS-33, BA-153 ^^UJI di >Vlj *5li U* a^i 

154a AS-33 r ^Jl j *>^j! V 

155a AS-33 JUJI o- v > j*j 

156a AK-56, AS-33 -criJu JyJI r M ^^ 

1 57a AK-56, AS-33 JT s3 j^l 


Al-Ghazdli on Divine Predicates 

158a AS-34 jU« * v^ **?& 
159a AK-57 \ } >X. otfj 
160a AS-34, BA-154 4Ub *Jy && 
161a AK-57 ^Ui r J 
162a AS-34 r,jj£, ^ju** ^01 jaJXi 
=* BA-154f J >*t..- v ^tj|.-. 
163a*AK-57 f/ J o& vJ=T Jul y, 
164a AK-57 ^j i^I ^Vl 6^T ^a^j r Jj 
165a AS-34 OiCl Jtf *j j>« Ul a^j 
166a AK-57 U**- Tj^ fjdi 
167a AK-57 *) Js-jU l-U 
168a AS-34. BA-154 «J^l j* 
169a AK-58, AS-34, BA-155 >>~yX *JI3 J* VI Jju V ( 
170a AS-34 f Sk\\ J* Jjlj *itf ^JOJir j 
171a AK-58 JUfeT j 
172a AK-58 >*?& j 
173a AK 58, BA-155 ,J«J»j 
174a AS-35. BA-155 f £*l **s 
175a BA-155, AK-58 tfjU-di J^t ^j ^i ^^ & 



Allard, M., Le probleme des attribute divim, Beyrouth : Imprimerie 

catholique, 1965. 
Aristotle. "Analytica Posteriora," WorAs, Vol. I (transl. G.R.G. 

Aristotle, "Metaphysica." The Works of Aristotle. Vol. VIII 

(transl. W.D. Ross). Oxford : Clarendon Press. 1960. 
Aristotle. "Physics," Works, Vol. II (transl. R P. Hardie and 

RK.Gaye). ' , 

Aristotle. "Topica," The Works of Aristotle, Vol. I (transl. 

W.A. Pickard-Cambridge), Oxford : Clarendon Press, 1955. 
al-Ash'ari. Abu al-Hasan. "Kitab al-Ibanah," Al-Rasd'il aU 

Sab'ah, Hyderabad, 1968. 
al-Ash'ari, Abfi al-Hasan, Maqaldt al-Isldmtyyln (ed. H. Ritter), 

Wiesbaden : Franz Steiner Verlag Gmbh, 1963. , 

al-Baghdadi. Abu Mansur 'Abd al-Qahir b. Jahir b. Muhammad, 

al-Farq Bayn al-Firaq, Cairo, 1910. 
al-Baqilani, al-lmam Abu Bakr, AU Tarn hid, Cairo : Dar al-Fikr, 


al-Dawwani. Shark al-'Aqd<id aW Adadiyyah. Cairo, 1322 ah. 

al-Fakuri. Hanna and al-Jarra. Kbalil. Tdrikh al-Falsafak aU 
•Arablyyah, Beirut : Dar al-Ma'arif, 1957. 

al-Farabi, Abu Nasr, Ard' akl aUMadlnak. Cairo, 1948. 

al-Farabi, Aba Nasr, Kitdbal-Jam' bayn Ra'yah aUHdkmayn (ed. 
A.N. Nadir). Beirut. 1960. 

al-Farabi. Abu Nasr. Kitab as-Siydsat al-M*danlyyak (ed. 
F.M. Najjar). Beirut. 1964. 

al-Ghazali, al-lmam AbQ Hfimid b. Muhammad. Ihyd' 'Ulutn ad- 
Din Cairo: La jnatNashral -ThaqafahaMslaroiyyah. 1356a B. 

al-Ghazali, al-lmam Iqtisdd pi-l'tiqid (eds. H. Atay and I. 
Cubuhcu), Ankara: Ankara Universltesi Ilahiyat Fakultesi 
Yavinlari, 1962. ' _ 

al-Ghazafi, aUmam, Jawdkir al-Qur'in, Cairo : Kurdistan Press, 

1329 ah. 


Al-Gkai&li on Divine Predicates 

al-Ghazali, al-Imam, Maqasid al-FalSsifah (ed. S. Dunya), Cairo, 

al-Ghazali, al-Imam, Al-Qist&s al-Mustaqim (ed. al-Rabbani ad- 

Dimashqi), Cairo : Mafha'at at-Taraqqi, 1900. 
al-Ghazali, al-Imam, Tahdfut al-Fal&sifah (ed. S. Dunya), Cairo, 

Ibn Hazm, al-Imam Abi Muhammad al-£ahiri, Kit&b al*Fi$alfi al 

Milal wa aUAhwd* wa al-Nthd, Baghdad : Muthanna, n.d. 
Ibn Rushd, Muhammad b. Ahmad, ManihijaUAdillah, dairo, 1935, 
Ibn Sina, Abu 'AH, AUIshdrdt wa al-Tanbihat (ed. S. Dunya), 

Cairo, 1958. 
Ibn Sina, Abu 'All, Al-Najil fi al-Hikmah al-Mantiqlyyah wa al- 

TaU'ah al-ll&hlyyah (ed. al-Kurdi), Cairo, 1938. 
Jabre, Farid, La notion de certitude selon GhazaM, Paris : Librairie 

philosophique, 1958. 
Khayyaf, Abfi al-Husayn *Abd al-Rahim, Kit&b al-Intisir wa al- 

Radd 'ala ibn al-Rawandi al-Mulhid, Cairo : Imprimerie de la 

Bibliotheque Egyptlenne, 1925. 
MacDonald, D.B., "Continuous Re-creation and Atomic Time," 

ISIS, No. 9 (1927). 
al-Murtada, Ahmad ibn Yahya, KiUb Jabaqdt aUMu'taxilah (ed. 

S. Diwald-Wilzer), tfeirut, 1961. 
al-Maturidi, Abi Mansfir, "Sharb al-Fiqh al-Akbar," Rasa'il al- 

Sab'ah, Hyderabad :Da'irat al-Ma'arif al-'Uthmaniyah, 1948. 
Pines, E., Ma&hhab al-Dharrah ind aUMuslimin (transl. Muham- 
mad 'Abd al-Hadl, Abu Rldah). Cairo, 1946. 
Plato, "Phaedo," The Dialogues of Plato, Vol. I (transl. B. Jowett), 

Oxford : Clarendon Press, 1953. 
Plato, "Republic," "Parmenides." ibid, Vol. II. 
Plato, "Sophist." "Timaeus," "Phaedrus," ibid., Vol. III. 
Prtzel, O., "Die Friihislamische Atomeulehre, " Der Islam, Vol. 19 


ai-Qagl, *Abd al-Jabbar, Shark d-Vstl aUKhamsah (ed. A. 
♦Uthman). Cairo, 1965. 

al-Shahrastani, Al-Milal wa al-Nthal, Leipzig : Otto Harrassowitz, 

al-Taftazani, Sa'd al-Din, Shark al-'Aqi'id, 'ala Main al-Aqa'idliU 
Nasaji, Cairo, n.d. 



Watt, W. Montgomery, Muslim Intellectual, A Study of al-Ghazdli, 

Edinburgh : Edinburgh University Press, 1963. 
Wolfson, H.A., "Avicenna, Alghazali, and Averroes," Homenage A 

Millas-Vallicrosa, Vol. II, Barcelona, 1956. 
Wolfson, H.A , "Maimonides on Negative Attributes," Louis 

Simberg jubilee Volume. 
Wolfson, H.A., Philosophical Implications of the Problem of- 

Divine Attributes in the Kalam," Journal of the American 

Oriental Society, Vol. 79. 
Wolfson, H.A., "The Amphibolous Terms in Aristotle, Arabic 

Philosophy and Maimonides," Harvard Theological Review, -31 

Wolfson, H.A., "The Muslim Attributes and the Christian 

Trinity," ibid., 49 (1956) 



Abraham, 40 

Abu 'Abdullah Muhammad ibn 

Karram. 39 
Abu 'All al-Juba'i, 75 
Abu Hashim, 75, 76 
actuality. 45 
•adam, 82, 84, 85 
AM al-Jjam', 45 
AM al-Sunnah, 63 
akwil. IS, 76 
al-Akhtal, 64 
•Slim, 65, 68, 70 
'ilimiyyah, 4 

al-'Allaf, Abu-I-Hudhayl, 22, 26, 39 
amir, 62 
amir. 68 
"Amphibolous Terms in Aristotle, 

Arabic Philosophy and Maimo- 

nides," 64 
amr, 52, 69 
amt-nahy-khabar, 70 
amr wa nahy, 68 
angels, 8 

annihilation, act of, 39 
anthropomorphists, 46 
•Aqd'id of Nasa/i, 27 
Arabs, 5t 
.Aristotle, 64 
el-Ash'ari, 22 ff.. 63. 64 

Ash'arites, 21, 23, 64, 95 

attributes, problem of the peculiar- 
ity of, 34 

"Avicenna, Alghcucal and Aver- 
roes," 22 

Avicenna's Psychology. 45 

«r«, 60 

al-Baghdadi, al-Shaykh Abu al- 

BarakSt, 27, 76 
Uqi, 98 
baqiyak, 83 

beatific vision, 45 

Beirut, 76 

belief, golden mean in, 10 

Bible, 70 

bi nafsiki, 73. 76 

Bishr ibn al-Mu'tamar, 22 

Book, 13 ; see also Qur'in 

"Categories" (oi Aristotle), 64 

causality, doctrine of, 21 

changeable-material-particulars, 28 

consensus, 47, 54, 57, 62, 70, 81 

contingent things. 3, 20, 26, 33, 36, 
37, 39, 41, 48, 49, 77, 79 «., 87 ft. 

Corporealists, 39 

creation, act of, 39 ; contingent, 39 ; 
doctrine of continuous, 24 

creatures, more deserving of limita- 
tions and weakness, 17 

daf'ah, 24 

darb, 51 

al-Dawwani, 27. 28 

"Destruction of the Philosophers," 

34, 37; see also Takdfut al-Fald- 

devils, 8 
dhdt, 2, 32, 57. 65, 67, 69, 74, 79; 82, 

83, 85. 86. 100 
dhihdb, 24 
Dirir, 24 
Diwald-Wilzer, Susanna, 76 

essence, 69. 71, 8\ 89; eternal, 34, 
82; God's. 32. 33. 36. 55. 57 ff.. 62, 
65, 76 ff.. 81. 83, 87. 89 ff., 93, 95, 

Eternal, 33, 58.65, 80 

eternity. 60, 82 

existence. 92, 96, 99 

existent, existent, 68 ; knowing, 68 



ai-Fargbi. 39 
fihr, 50 
fl'l. U 

first creature, 79 
freedom, man's, 21 
free will, 23 

Gabriel. 63 

generation, 17 ff. 24 

gkajlak, 84 

gkani, 98 

GhasaU 21 ff., 29, 39, 45, 46, 76 

golden mean, 10, 71 

Great Day of Fright. 93 

kdditk, 27, 32, 36, 59, 82, 87 

hoditkak, 77 

kaditku-n-nafs, 49. SO 

kdlak. 65, 86 ' 

Mmqiqak. 23 

Harvard Tkeological Jtevuw, 64 

kayak, 86 

Homenaj* A At ilia* I alliiro**, 22. 7o 

honeycombs, V 

"bowness." 22 

lba Rushd, 29, 39 
ibn Sina, 23. 23. 39. 45. 76. 97 
al-i'dam, 39 
idolaters, 62 
at ikdatk, 39 

al-ijad, 39. 87, 91 ;— haditk, 91 
ijmd', 40; sec also consensu* 
ikhtira'o, 10 
ikktiydr, 23 
al-iktisdb, 24 
Urn, 6, 30. 31, 36. 41, 62, 67 fl, 83 ; 

see also knowledge 
imk&n. 3, 5, 6 
incorporeal beings, 22 
IntisAw, 21. 23 
Iqtisid 45 
irddah, 35, 78, 87. 91 ; —la (i miHu i 

lihardt, 2.1 

al-iskafl. 24 
Islam. 60, 74 
istikbdr. 51 

jabr, 10 

Jabrites. 9. 23. 97 

Jahm ibn Safwan, 23. 8J, 87, 90, 97 

al-Jabmiyyab, 97 

ja'iz, 5 

j a wad. 99 

jawihir, 69 

jawj, 17 

jawkar, 4, 75 

Jinn, 8 

journal of the American Oriental 

Society, 22 
al-Juba I, see Abu 'All 
jurisprudence, 74, 75 

ai-Ka'bvAbu-l-jjasim ibn Muham- 
mad. 30, 39. 7o 

hatam, 93 tt\ 87, 93 

kaldmu-n-nafi. 50, 63. 64, 7.1 

al-K*riauii, VI 

Karramites, 39, 7|. 76, 83, 87. 91, 97 

kasb, 13, 22 ff. 

kayjiyyh 22. 57 

ttkabar, 51. 68 

kkdliq, 98 ff 

ul-Khayyat, 39 

Kitdb an-Najs, 2v 

Kildb fabqdt al-Mtftazilah. 76 ' 

knowledge. IS, 20, 30. 31. 36. 37, 
40 if.. 50. 51, 54. 62. 64. 65, 67 ff., 
74, 78, 83 ff., 94, 96; God's eternal] 

U-dhdlihi. 76 

Law, Divine, 8, 58. 7f, 74, 75 

natural, 4 
idtimdt. V) 
life, and deatb. pow«r ou. 22 ; crea- 

tiefa of, 7 
likcnen 34 
Louis Gimztberfi fubilte Volume. 22 


Al-Ghazali on Divine Predicates 

isktitdk, 38 

aUMabahUk al-Maskriqiyyak, 27 

Afddkkab aldhdrrah.23 

makal (locus), 33 

"Maimonides on Negative Attri- 
butes," 22 

majbur, 23 

dm Mm, 36 

Ma'mar. 24 

Mandkij al-Adtlluh. 22. 3*> 

Maqaldt. 22 II. 27. 97 

Maqru', 58 if. 

masdAt/. 57. 58. 64 

Maturidi, 23, 64 

mawjud, 68 

Messenger, 47. 60, 61 ; «« «•'« 

at-Milal wa al-Nihal.. 39. 63. 64.-97 

moderation, 70 

Muses, 54 6,62. 87, 93 

motion, 11. 12. 32 78.84 

movement. 4. 7, 12. 17 li., 22 

mubdshir, 24 

mudkil, 98. 99 

mufti 53. 54 

muhil, 4 8 . IK- 23 

MwAiolA, 38 

mn'w.98. 99 

iMW>oMtmaA. 46 

mujid, 39 

m«**Wr. 69 

, MM m*»«. 4 0., 23. 81 

MMfNtani', 23 

murajjik, 23. 39 

Musi, 54 

muskaf. 5 ? . 58 
mtuA'**. 62 

tmusklarak. 64 

Muslims. 45, 47 

mutaharnkah. 85 

Mutakalltms, 29 

mulawathd, 24, 64 

Mutaxilah, 8. 16, 20. 22f. 26. 27. 2V. 

.13. 35. 38. 39, 41, 46. 63, 65, 07. 69. 

naky. 69 

al-Najit, 29. 46. 76 

al-Najjar, 24 

n&r (Are), 58 

Naznm, 21. 22. 26 

Necessary Being. 80, 81 

Necessary Essence, 80 

Noah, 87, 93 

iion-rxutrucr. 83, 80. 92. 96. 98 

obstinacy, 2 
omnipotence, 10 

particulars. 27. 28 

People ui tbc Truth. 30 

periection 42, 43, 45. 48. 49; wonders 

of, 2 

'Philosophical Implications of the 
Problem of Divine Attribute* in 
the Kalain," 22 

Piues. S . 23 

poles. 34 

power (qudrah). 1, 2. 3. 7. 11 ff , 16 
II . 21. 23 it.. 37, 49. 51. 65. 6K, 71. 
73, 8U, 86. 87. 93, 96. 100 ; contin- 
gent. 16. 20, 21. 24 ; Divine, 21, 74; 
eternal, 51. 32; God's absolute, 21; 
object* of. 3 ; relation between, 
and its objects, 3, 13 ; relativity 
of. 15 

powerlcssness. 7, 21 

predestination {jabr), doctrine of, 10 

Prophet. 66. 73. 74 ; set also Messen- 

qabik, 38 

qadim.l, 3. 5. 61.83. 86.91,95.98 
qadimuk, 3 
jl.qadiriyal-ulqudimak, 16 

qalb, 50 
bdri', 5y 
qasd, 24 

qttu'uh, SS li 




n&kin, 68 

quwwak mufakkirak 50 

Rafidis, 11 
UatuiMM, i\. 45 
id-Kazi. Imam, 27 
riiiq. 98, 99 
rest. 7. 12. 22, 84 

sdkinak, 85 

al-Salibl. 22 

separateness, 86 

Shdhrastaol. 39 

skar\ 58, 70, 75 

Skark al-'Aqd'id al-Adadiyyak, 27 

Shark at-Fiqk aUAkbat, 23. 64 

ih.nik, 98 

sildk al-ma'na,99 

speech. God's, 47. 48. 55 ft., 04, 69. 

74, 77. 80 ; eternal, S8 ; inner. 30 

54, 63, 73 93 
substances, 3, 4, 40. 22. 69 
Sufis, 45 
sultdn. 52, 53 
Sunnak, people of, 10 
Sunnls. 32, 33, 38, 63 

qudrak, 68, 74 ; set also power 
quiddities, 27 

Qur'an. 22. 23. 40. 57 fl„ 63. 64. 66. 
•i ablet," 63 

al-TafUzaiii, Sa'd al-Dlo. 27 
T aha jut al Faldsijah, 34. 35. 39 
tajit. 21 
lanl al-'adak, 21 
td'tkir, 13 

tawallud, 17, 20, 21, 24 
To rah, 70 
Tusi. 23. 28. 

Ummak. 42. 60 
universals, 27, 28 
•(Unman. 64 

Vision, God's, 46 

udjik, 5. 23. 97 

tPilihu-l-dh&t, 80 

wajibu-l-veujud, 80 

will, aud knowledge, relativity of, 
14 ; contingent. 35, 36 ; eternal, 
31 fi.. 36 fi. ; eternity of, 33 : exist- 
ence of, 5 ; lack of. 5 ; properties 
of, 3 

Wolfson, 22. 64, 76 

wujud, 85. 92